Smad proteins differentially regulate obesity-induced glucose and lipid abnormalities and inflammation via class-specific control of AMPK-related kinase MPK38/MELK activity

Smad proteins differentially regulate obesity-induced glucose and lipid abnormalities and... Smad proteins have been implicated in metabolic processes, but little is known about how they regulate metabolism. Because Smad 2, 3, 4, and 7 have previously been shown to interact with murine protein serine–threonine kinase 38 (MPK38), an AMP‐activated protein kinase (AMPK)-related kinase that has been implicated in obesity-associated metabolic defects, we investigated whether Smad proteins regulate metabolic processes via MPK38. Smads2/3/4 increased, but Smad7 decreased, MPK38-mediated apoptosis signal-regulating kinase-1 (ASK1)/transforming growth factor-β (TGF-β)/p53 signaling. However, MPK38-mediated phosphorylation-defective Smad mutants (Smad2 S245A, Smad3 S204A, Smad4 S343A, and Smad7 T96A) had no such effect. In addition, Smads2/3/4 increased, but Smad7 decreased, the stability of MPK38. Consistent with this, Smads2/3/4 attenuated complex formation between MPK38 and its negative regulator thioredoxin (Trx), whereas Smad7 increased this complex formation. However, an opposite effect was observed on complex formation between MPK38 and its positive regulator zinc-finger-like protein 9 (ZPR9). When Smads were overexpressed in high-fat diet (HFD)-fed obese mice using an adenoviral delivery system, Smads2/ 3/4 improved, but Smad7 worsened, obesity-associated metabolic parameters and inflammation in a MPK38 phosphorylation-dependent manner. These findings suggest that Smad proteins have class-specific impacts on obesity-associated metabolism by differentially regulating MPK38 activity in diet-induced obese mice. Introduction cell proliferation, differentiation, apoptosis, migration, The identification of a growing list of intracellular extracellular matrix remodeling, immune functions, and kinases that phosphorylate Smad proteins suggests that tumor metastasis. This occurs through the combined use the transforming growth factor-β (TGF-β)/Smad signaling of TGF-β signaling pathway components, such as Smads pathway cross-talks with a variety of other intracellular and Smad-interacting transcription factors, cross-talk signaling pathways . The TGF-β signaling pathway reg- with other intracellular signaling pathways, and the ulates a broad range of cellular processes, which include ability of TGF-β receptors to activate other 2–6 signaling modules . Many studies have shown that Smads are phosphorylated by multiple intracellular kina- ses, including mitogen-activated protein kinases, Correspondence: Hyunjung Ha (hyunha@cbnu.ac.kr) 1 2+ Department of Biochemistry, School of Biological Sciences, Chungbuk Ca /calmodulin-dependent kinase II, cyclin-dependent National University, Cheongju 28644, Republic of Korea kinase (CDK), protein kinase C, G protein-coupled Department of Biochemistry, University of Madras, Guindy Campus, Chennai receptor kinase 2, extracellular signal-regulated kinase, 600025, India Edited by C. Munoz-Pinedo © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; Seong et al. Cell Death and Disease (2018) 9:471 Page 2 of 13 apoptosis signal-regulating kinase-1 (ASK1), and murine molecular mechanisms involved in the regulation of meta- protein serine–threonine kinase 38 (MPK38)/maternal bolic homeostasis by TGF-β signaling remain poorly 1,7,8 embryonic leucine zipper kinase (MELK) , suggesting understood. that the TGF-β pathway is closely integrated with other In this study, we show that there are direct physical and intracellular signaling pathways to achieve tightly regu- functional interactions between MPK38 and Smads lated TGF-β responses. However, most of these studies (Smad2, 3, 4, and 7). Smads2/3/4 stimulate MPK38- dependent ASK1/TGF-β/p53 signaling pathways, whereas have focused on the regulatory role of Smad phosphor- ylation in the TGF-β signaling pathway. Additional stu- Smad7 inhibits these signaling pathways through differ- dies are required to investigate the effect of Smad proteins ential regulation of MPK38 activity. Furthermore, over- on the activity of these interacting kinases in order to expression of Smads2/3/4 improves, whereas Smad7 decipher the molecular interplay between TGF-β and overexpression worsens, obesity-associated metabolic other intracellular signaling pathways. parameters by differentially regulating MPK38 activity in MPK38/MELK, an AMP‐activated protein kinase HFD-induced obese mice. (AMPK)-related kinase, has been shown to mediate var- ious cellular functions, including proliferation, spliceo- Results some assembly, gene expression, carcinogenesis, MPK38 kinase activity is increased by Smads2/3/4 but 9–13 apoptosis, and metabolism , although its exact phy- decreased by Smad7 siological functions still remain to be determined. MPK38 Given that MPK38 interacts with and phosphorylates and its interacting partner Smad3 have recently been Smad proteins, leading to the activation of TGF-β signal- shown to serve as components of a multi-protein complex ing , we reasoned that Smad proteins would affect MPK38 linking ASK1 and TGF-β signaling pathways, which are activity through direct interaction and phosphorylation. To involved in glucose and lipid metabolism in mice, and to assess this possibility, we performed immunoblot analysis contribute to the activation of ASK1 signaling via a direct using CRISPR/Cas9-mediated Smad knock-in (KI) HEK293 8,11 interaction with ASK1 . TGF-β1 was previously repor- cells (Smad2 S245A, Smad3 S204A, Smad4 S343A, and ted to positively regulate the 3-phosphoinositide- Smad7 T96A), which are defective in MPK38-mediated 14 7 dependent protein kinase-1 (PDK1)/AKT1 pathway , phosphorylation , to determine the endogenous kinase although PDK1 was shown to inhibit TGF-β signaling activity of MPK38 in the presence or absence of ASK1/ through direct interactions with Smads . These findings TGF-β/p53 signals, including H O,TGF-β1, and 2 2 suggest potential roles of Smads in the regulation of key 5-fluorouracil (5FU). The endogenous kinase activity of kinases involved in intracellular signaling pathways that MPK38 was markedly lower in the S245A, S204A, and are integrated with TGF-β signaling. S343A KI cells when compared with wild-type control cells, Recent discoveries have shed some light on the important whereas T96A KI cells had higher MPK38 kinase activity role that TGF-β signaling plays in adipose physiology and (Fig. 1a). These results were corroborated by inhibition of 16–18 metabolism . Smad3 deficiency in mice resulted in the kinase activity of MPK38 using a potent MPK38 inhi- improved glucose tolerance and insulin sensitivity, accom- bitor OTSSP167 (Fig. 1a) . We also analyzed the effects of panied by reduced white adipose tissue (WAT) mass and phosphorylation of Smad isoforms by MPK38 in the reg- browning. The associated increase in mitochondrial bio- ulation of MPK38 kinase activity using in vitro kinase assays genesis resulted in the dissipation of the excess energy with recombinant wild-type and MPK38-mediated phos- 16,17 stored in WAT by thermogenesis .Higher TGF-β1in phorylation-defective Smad mutants. The MPK38 kinase humans has been shown to positively correlate with greater activity was increased by recombinant wild-type Smads2/3/ adiposity and a poor metabolic profile, and to negatively 4, but decreased by recombinant wild-type Smad7. How- correlate with fitness . Several recent studies have ever, the recombinant MPK38-mediated phosphorylation- demonstrated that TGF-β signaling regulates insulin gene defective Smad mutants had no such effects (Fig. 1b). These transcription in pancreatic β cells . Moreover, the Smad3 findings indicate that Smad proteins differentially regulate gene was identified in a genome-wide association study for the kinase activity of MPK38, resulting in specificSmad- type 2 diabetes risk .These findings implicate Smad3 as a mediated regulation of coordinate MPK38-induced ASK1/ potential target for the treatment of obesity and its asso- TGF-β/p53 signaling. ciated disorders. Conversely, targeted disruption of Smad2 in mouse pancreatic β cells caused islet cell hyperplasia and MPK38-dependent ASK1/TGF-β/p53-mediated impaired insulin secretion by attenuating ATP-sensitive K+ transcription, apoptosis, and signaling activation are channel activity . However, inhibition of Smad4 in pan- increased by Smads2/3/4 but decreased by Smad7 creatic β cells conferred minor but significant improve- Because MPK38 has been shown to potentiate coordi- 7,11,23 ments in blood glucose and glucose tolerance in high-fat nate ASK1/TGF-β/p53 signaling ,we first examined diet (HFD)-induced obese mice . Nevertheless, the the effect of Smad proteins on ASK1/TGF-β/p53- Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 3 of 13 Fig. 1 (See legend on next page.) Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 4 of 13 (see figure on previous page) Fig. 1 Differential regulation of MPK38 kinase activity by Smad proteins. a To assess the effects of wild-type (WT) and MPK38-mediated phosphorylation-defective Smad mutants (Smad2 S245A, Smad3 S204A, Smad4 S343A, and Smad7 T96A) on MPK38 kinase activity, WT and CRISPR/ Cas9 Smad knock-in HEK293 cells (S245A, S204A, S343A, and T96A) were incubated with (+) or without (−) OTSSP167 (1 μM, 2 h), a potent MPK38 inhibitor, and then treated with or without the following stimuli: H O (2 mM, 30 min), TGF‐β1 (2 ng/ml, 20 h), or 5FU (0.38 mM, 30 h). The cell lysates 2 2 were subjected to immunoprecipitation with antibodies for ZPR9, STRAP, and Trx, followed by immunoblot analysis using anti-phospho-specific 23,24,34 252 188 76 antibodies for ZPR9 Thr , STRAP Ser , and Trx Thr . The protein levels of MPK38 and Smads in cell lysates were examined with anti-MPK38 and anti-Smads antibodies. b Recombinant MPK38 proteins (~5 μg) were screened in in vitro kinase assays using recombinant ZPR9 (~5 μg) as a substrate in the presence of recombinant WT and MPK38-mediated phosphorylation-defective Smad mutants (~0.8 μg each). Ratios of P-ZPR9/ZPR9 32 32 a, P-ZPR9/ZPR9 b, and P-MPK38/MPK38 b were determined by density analysis of the bands, and the fold-increase relative to the untreated WT 32 32 controls a or control lacking Smad expression b is presented for each protein. P, P incorporation; P, phosphorylated; re., recombinant mediated transcription induced by MPK38. As shown in Smads (Smad2, 3, 4, and 7) on MPK38 protein stability. Fig. 2 and Supplementary Fig. 1, the ASK1/TGF-β/p53- HEK293 cells were transfected with expression vectors mediated transactivation induced by MPK38 increased in encoding wild-type and MPK38-mediated phosphoryla- a kinase-dependent manner in the presence of wild-type tion-defective Smad mutants, and MPK38 protein levels Smads2/3/4, whereas MPK38-mediated phosphorylation- were quantified using immunoblot analysis. As shown in defective Smad mutants had no such effect. By contrast, Fig. 3a, Smads2/3/4 increased the stability of MPK38 Smad7 decreased the transcriptional activity. These compared with control cells expressing empty vector, results suggest that Smads2/3/4 are positive regulators of whereas Smad7 decreased MPK38 stability. However, MPK38 and that Smad7 is a negative regulator of MPK38. such effects were not observed in the presence of MPK38- The data also demonstrate that the phosphorylation of mediated phosphorylation-defective Smad mutants. Smad proteins by MPK38 is critical for specific Smad- Conversely, treatment of Smad-expressing HEK293 cells mediated regulation of ASK1/TGF-β/p53-mediated with both cycloheximide and MG132, a proteasomal transactivation induced by MPK38. inhibitor, led to greater stability of MPK38, compared We next examined whether Smad proteins can influence with that in non-MG132-treated Smad-expressing ASK1/TGF-β/p53-mediated apoptosis induced by MPK38. HEK293 cells (Fig. 3a). These results indicate a critical As shown in Fig. 2 (middle panels) and Supplementary role for the proteasome pathway in the effects of Smads Fig. 1, the coexpression of wild-type Smads2/3/4 and on MPK38 degradation in cells. We then analyzed the MPK38 resulted in a kinase-dependent increase in ASK1/ effect of Smads on MPK38 ubiquitination using CRISPR/ TGF-β/p53-induced apoptosis compared with control cells Cas9-mediated Smad KI HEK293 cells treated with (+)or transfected with MPK38 alone. However, the stimulatory without (−) OTSSP167. Smads2/3/4 endogenously effect of Smads2/3/4 on ASK1/TGF-β/p53-induced apop- decreased the ubiquitination of MPK38 in a MPK38 tosis was not observed in the presence of MPK38-mediated phosphorylation-dependent manner, whereas Smad7 phosphorylation-defective Smad mutants. Conversely, the increased MPK38 ubiquitination (Fig. 3b). These results opposite effect was observed in the presence of Smad7. suggest that, in addition to the class of Smad protein, We also evaluated MPK38-mediated ASK1/TGF-β/ MPK38-mediated phosphorylation of Smad proteins also p53 signaling activation using CRISPR/Cas9-mediated plays a critical role in the regulation of MPK38 protein Smad KI HEK293 cells in the presence or absence of H O / stability. 2 2 TGF-β1/p53. As shown in Fig. 2 (right panels), and as We then investigated whether the specific Smad- expected, the activation of ASK1/TGF-β/p53 signaling was mediated regulation of MPK38 stability is dependent on markedly lower in S245A, S204A, and S343A KI cells when the interaction between MPK38 and Mdm2 . HEK293 compared with wild-type control cells, whereas T96A KI cells were transfected with expression vectors for wild- cells showed greater activation of ASK1/TGF-β/ type and MPK38-mediated phosphorylation-defective p53 signaling. Together, these findings suggest that Smad mutants. Smads2/3/4 markedly decreased MPK38- MPK38-induced ASK1/TGF-β/p53 activity is positively Mdm2 complex formation in a MPK38 phosphorylation- regulated by Smads2/3/4, but negatively regulated by dependent manner, whereas Smad7 increased complex Smad7, in a MPK38 phosphorylation-dependent manner. formation (Fig. 3c). These results also indicate the important role of MPK38-mediated phosphorylation of MPK38 stability is increased by Smads2/3/4 but decreased specific Smads in the differential regulation of Mdm2- by Smad7 dependent MPK38 ubiquitination. All these results sug- To address the mechanism of the differential regulation gest critical roles for specific Smad protein classes and of MPK38 activity by Smads, we investigated the effect of their phosphorylation by MPK38 in the regulation of Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 5 of 13 Fig. 2 (See legend on next page.) MPK38 stability: Smads2/3/4 stabilize MPK38, whereas siRNAs were subjected to immunoprecipitation using an Smad7 reduces its stability. anti-MPK38 antibody, followed by immunoblot analysis using an anti-Trx antibody. There was a dose-dependent Smad proteins differentially regulate complex formation decrease in endogenous complex formation between between MPK38 and its regulators, thioredoxin (Trx) and MPK38 and Trx in cells expressing Smads2/3/4 compared zinc-finger-like protein 9 (ZPR9) with cells not expressing Smads. By contrast, Smad7 To further explore the mechanism of differential reg- transfection increased complex formation (Fig. 4a). These ulation of MPK38 stability by Smads, we investigated results were corroborated by Smad silencing experiments whether Smad proteins affect the interaction between using Smad-specific siRNAs (Fig. 4a). These findings MPK38 and Trx, which destabilizes it . HEK293 cells indicate that Smad class type contributes to the Smad- transfected with various quantities of plasmid vectors mediated differential modulation of MPK38-Trx complex encoding Smads2/3/4/7 or specific Smad-targeting formation. Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 6 of 13 (see figure on previous page) Fig. 2 Differential regulation of MPK38‐mediated ASK1/TGF-β/p53 signaling by Smads. a 293T cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg), WT and K40R MPK38 (0.8 μg), WT ASK1 (0.6 μg), c‐fos (0.6 μg), or the AP‐1- luciferase plasmid (0.2 μg), as indicated. After 48 h, the cells were harvested, and luciferase activity was measured using a luciferase assay system (Promega). The total DNA concentration was kept constant by supplementation with empty vector DNA. The values were adjusted relative to the expression levels of a co-transfected β-galactosidase reporter control (left panels). HEK293 cells were transfected with various concentrations of vectors encoding WT and mutant Smads3/7 (0.6 and 1.2 μg), or WT and K40R MPK38 (0.8 μg), as indicated, in the presence or absence of H O (1 mM, 2 2 9 h). Cells exposed only to H O were used as a positive control. Apoptotic cell death was determined using a GFP system (middle panels). WT and 2 2 CRISPR/Cas9 Smad knock-in HEK293 cells were treated with (+) or without (−)H O (2 mM, 30 min), and the cell lysates were subjected to 2 2 immunoprecipitation with antibodies for ASK1, MKK3, p38, and ATF2, followed by immunoblot analysis using anti‐phospho-specific antibodies for 845 838 189/207 180 182 71 ASK1 Thr (Thr in human), MKK3/6 Ser , p38 Thr /Tyr , and ATF2 Thr . The protein levels of MPK38 and Smads in cell lysates were examined with anti-MPK38 and anti-Smads antibodies (right panels). *p < 0.05, **p < 0.01, ***p < 0.001 compared with MPK38 alone in the presence of ASK1 or H O . b HaCaT cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg), WT and 2 2 K40R MPK38 (0.4 μg), or p3TP‐Lux plasmid (0.2 μg), as indicated, in the presence or absence of TGF‐β1 (100 pM) (left panels). HaCaT cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg) and/or WT and K40R MPK38 (0.4 μg), as indicated, together with an expression vector encoding GFP (1 μg). After treatment of the transfected cells with TGF‐β1 (2 ng/ml, 20 h), apoptotic cell death was then determined (middle panels). WT and CRISPR/Cas9 Smad knock-in HEK293 cells were treated with (+) or without (−) TGF‐β1 (2 ng/ml, 20 h), and the cell lysates were analyzed by immunoblot analysis using antibodies for PAI-1, p21, Smad7, CDK4, cyclin D1, MPK38, and Smads (Smad2, Smad3, Smad4, and Smad7) (right panels). **p < 0.01, ***p < 0.001 compared with MPK38 alone in the presence of TGF-β1. c MCF7 cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg), WT and K40R MPK38 (0.4 μg), or p53-Luc plasmid (0.2 μg), as indicated, in the presence or absence of p53 (0.3 μg) (left panels). MCF7 cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg) and/or WT and K40R MPK38 (0.4 μg), as indicated, together with an expression vector encoding GFP (1 μg) in the presence or absence of p53 (0.6 μg). Apoptotic cell death was then determined (middle panels). WT and CRISPR/Cas9 Smad knock-in HEK293 cells were treated with (+) or without (−) 5FU (0.38 mM, 30 h), and the cell lysates were analyzed by immunoblot analysis using antibodies for p53, p21, Mdm2, Bax, MPK38, and Smads (Smad2, Smad3, Smad4, and Smad7) (right panels). ***p < 0.001 compared with MPK38 alone in the presence of p53. The results represent the mean ± S.E. of at least three independent experiments performed in duplicate. Kinase-dead MPK38, K40R Because ZPR9 was recently shown to function as a adipocytes, compared with uninfected HFD-fed mice or stabilizer of MPK38 , we next investigated the effect of HFD-fed mice infected only with Green Fluorescent Smads on ZPR9 binding to MPK38. As expected, Smads2/ Protein (GFP)-expressing adenovirus (Ad-GFP) (Fig. 5a; 3/4 increased endogenous complex formation between Supplementary Fig. 2a). The mRNA expression levels of MPK38 and ZPR9, but Smad7 decreased this (Fig. 4b). key adipogenic regulators, including CCAAT-enhancer- Consistent with this, endogenous Smad silencing using binding protein α, peroxisome proliferator-activated Smad-specific siRNAs had the opposite effect on MPK38- receptor γ (PPARγ), and fatty acid binding protein 4, ZPR9 complex formation (Fig. 4b). These observations were significantly lower in WAT from HFD-fed mice provide evidence that the differential regulation of infected with Ad-Smads2/3/4 than those in the uninfected MPK38 stability by Smads is mediated by modulating HFD-fed mice. However, Ad-Smad7 infection had an complex formation between MPK38 and its regulators, opposite effect (Fig. 5b). Trx and ZPR9. Fasting blood glucose was lower in HFD-fed mice infected with Ad-Smads2/3/4 than in uninfected HFD-fed Smads2/3/4 improve, but Smad7 worsens, glucose mice (Fig. 5c) and Ad-Smads2/3/4 infection enhanced metabolism in a mouse model of diet-induced obesity glucose tolerance and insulin sensitivity in HFD-fed mice Emerging evidence has implicated ASK1/TGF-β/ (Figs. 5c, d). In parallel, mice infected with Ad-Smads2/3/ p53 signaling pathways in the pathogenesis of obesity- 4 exhibited lower circulating insulin levels under fasting 8,25,26 associated metabolic diseases . Our recent studies conditions, whereas Ad-Smad7 infection further have shown that both genetically and diet-induced obese increased the levels of circulating insulin (Fig. 5d, left). In mice display lower levels of ASK1/TGF-β/p53 signaling addition, Ad-Smads2/3/4 infection significantly increased and Smad3 expression, and a higher level of Smad7 in vitro insulin-stimulated 2-deoxy-glucose uptake in expression, when compared with control wild-type or WAT and muscle, whereas Ad-Smad7 infection had the 8,23 chow-fed mice . Based on these findings, an adenoviral opposite effect (Fig. 5e, left; Supplementary Fig. 2b). delivery system was employed to investigate whether Consistent with this, we observed that Ad-Smads2/3/4 Smad proteins regulate obesity-associated glucose meta- infection caused a significant upregulation of the insulin bolism by differentially regulating ASK1/TGF-β/ receptor substrate (IRS)-phosphoinositide 3-kinase (PI3K) p53 signaling in diet-induced obese mice. The pathway, which may contribute to enhanced glucose adenoviral delivery of Smads2/3/4 significantly decreased uptake, whereas the IRS-PI3K pathway was down- cellular distribution toward extremely large, hypertrophic regulated by Ad-Smad7 infection (Fig. 5e, right; Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 7 of 13 Fig. 3 Differential regulation of MPK38 stability by Smads. a MPK38 protein stability was assessed by immunoblot analysis using an anti‐MPK38 antibody. HEK293 cells were transfected with pCMV2‐FLAG (vector) or vectors encoding FLAG-tagged WT or mutant Smads. Time intervals indicate the number of minutes after treatment with cycloheximide (CHX, 20 μg/ml) alone or with MG132 (10 μM). The MPK38 levels in at least three independent experiments were quantified by densitometry (right panels). The relative densitometry at each time point was expressed as a percentage of the density at time 0 min after normalization the corresponding β-actin level. b The ubiquitination of endogenous MPK38 was assessed using WT and CRISPR/Cas9 Smad knock-in HEK293 cells treated with (+) or without (−) OTSSP167 (1 μM, 2 h). c HEK293 cells were transfected with vectors encoding FLAG-tagged WT or mutant Smads, as indicated. Cell lysates were subjected to immunoprecipitation using an anti-MPK38 antibody (IP: α-MPK38) followed by immunoblot analysis using an anti‐Mdm2 antibody to determine the endogenous levels of MPK38‐ Mdm2 complexes. Ratio of Mdm2/MPK38 was determined by density analysis of the bands, and the fold-increase relative to control not expressing Smads is presented Supplementary Fig. 1b). HFD-fed mice infected with Ad- 3/4 infection significantly reduced the mRNA expression Smads2/3/4 displayed a considerable decrease in blood of adipose and hepatic lipogenic genes, including fatty glucose levels (Supplementary Fig. 2c), together with acid synthase (FAS), sterol CoA desaturase 1, and sterol lower mRNA expression levels of hepatic gluconeogenic regulatory element-binding transcription factor 1c, con- genes, including glucose-6-phosphatase (G6PC), phos- sistent with lower circulating free fatty acid levels, phoenolpyruvate carboxykinase-1 (PCK1), and peroxi- whereas Ad-Smad7 infection had the opposite effect in some proliferator-activated receptor γ coactivator 1α both tissues (Fig. 6a; Supplementary Fig. 3a). Consistent (PGC1α), when compared with uninfected HFD-fed mice with this, and in contrast to the effects of infection with (Fig. 5f). By contrast, no MPK38-mediated phosphoryla- Ad-Smad7, HFD-fed mice infected with Ad-Smads2/3/4 tion-defective Smad mutants had such an effect (Fig. 5; exhibited lower lipogenesis (Fig. 6a), liver triglyceride Supplementary Fig. 2). These results indicate that (Fig. 6a), circulating total cholesterol, high-density lipo- Smads2/3/4 improve, but Smad7 worsens, glucose meta- protein (HDL)-cholesterol, and low-density lipoprotein bolism in a MPK38 phosphorylation-dependent manner (LDL)-cholesterol (Fig. 6b). However, there was no such in diet-induced obese mice. effect in the presence of MPK38-mediated phosphoryla- tion-defective Smad mutants. These results suggest that Smads2/3/4 improve, but Smad7 worsens, lipid Smad phosphorylation by MPK38 is important in the metabolism and inflammation in a mouse model of diet- Smad-mediated differential regulation of lipogenesis. A induced obesity previous report demonstrated that cholesterol, fatty acids, We then examined whether Smad proteins regulate and modified lipids activate inflammatory pathways and lipogenic gene expression in HFD-fed mice. Ad-Smads2/ modulate the activity of leukocytes . Therefore, we also Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 8 of 13 Fig. 4 Smad-specific regulation of complex formation between MPK38 and Trx or ZPR9. HEK293 cells were transfected with various concentrations of vectors encoding GST-tagged Smads (0.5 and 1 μg) or Smad-specific siRNAs (100 and 200 nM), as indicated. Cell lysates were subjected to immunoprecipitation with an anti-MPK38 antibody (IP: α-MPK38), and endogenous complex formation between MPK38 and its regulators, Trx a and ZPR9 b, was assessed by immunoblot analysis using anti-Trx and anti-ZPR9 antibodies (upper panels). Ratios of Trx/MPK38 a and ZPR9/MPK38 b were determined by density analysis of the bands, and the fold-increase relative to controls not expressing Smads is presented (lower panels). Each experiment was repeated at least three times with similar results examined the expression of proinflammatory genes in (Fig. 6d). Indeed, WAT from HFD-fed mice infected with blood samples. Ad-Smads2/3/4 infection considerably Ad-Smads2/3/4 exhibited higher expression of key genes decreased serum proinflammatory proteins when com- involved in fatty acid oxidation, such as peroxisome pared with uninfected HFD-fed mice, whereas Ad-Smad7 proliferator-activated receptor α (PPARα), carnitine pal- infection increased the serum levels of proinflammatory mitoyltransferase 1, and acyl-CoA oxidase, as well as proteins (Fig. 6c). However, these effects were not lower levels of blood triglycerides and higher observed after infection with MPK38-mediated phos- isoproterenol-stimulated lipolysis (Fig. 6e; Supplementary phorylation-defective Smad mutants. These results indi- Fig. 3c). HFD-fed mice infected with Ad-Smads2/3/4 also cate that Smads2/3/4 have beneficial effects, whereas displayed substantial decreases in liver lipid accumulation Smad7 has an adverse effect, on lipogenesis and and triglyceride content (Fig. 6a, f). However, we observed inflammation. the opposite trend in HFD-fed mice infected with Ad- Ad-Smads2/3/4 infections also increased the mRNA Smad7. All these results indicate the positive roles of expression of adipose lipolytic genes, including hormone- Smads2/3/4 and the negative role of Smad7 in lipid sensitive lipase (HSL), adipose triglyceride lipase (ATGL), oxidation. and beta-3 adrenergic receptor (ADRB3), in a MPK38 Because abnormal mitochondrial fat oxidation is phosphorylation-dependent manner, whereas Ad-Smad7 associated with insulin resistance and impaired keto- infection decreased the expression of these genes (Sup- genesis , we also investigated whether Smad proteins plementary Fig. 3b). Consistent with this finding, Ad- differentially regulate hepatic ketogenesis. As expected, Smads2/3/4 infection stimulated hepatic fatty acid utili- Ad-Smads2/3/4 infection enhanced ketone body pro- zation by stimulating the oxidation of fatty acids via duction (Fig. 6g) and ketogenic gene expression (Sup- mitochondrial and peroxisomal β-oxidation pathways in a plementary Fig. 3d) under fasting conditions when MPK38 phosphorylation-dependent manner, whereas compared with uninfected HFD-fed mice, but Ad- Smad7 had the opposite effect on fatty acid utilization Smad7 infection had the opposite effect. We also Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 9 of 13 Fig. 5 Specific effects of adenoviral delivery of Smads on glucose metabolism in diet-induced obese mice. a Hematoxylin and eosin (H&E)- stained paraffin-embedded sections of epididymal WAT in HFD-fed mice infected with the indicated adenoviruses. Scale bar, 100 μm. b Relative mRNA expression levels of adipogenic genes. mRNA expression was quantified by densitometry, and the fold-increase relative to control is presented. n = 6 per group, *p < 0.05, **p < 0.01 compared with control. c, d Blood glucose c and insulin d levels in fed and fasting (16 h) HFD-fed mice that had been infected with the indicated adenoviruses or not (control) (left panels). n = 6 per group, *p < 0.05, **p < 0.01 compared with the fasted controls, determined by two-way ANOVA. Glucose tolerance tests c and insulin tolerance tests d were conducted by measuring blood glucose concentrations in mice following intraperitoneal injection of glucose (2 g/kg) or insulin (0.75 U/kg), respectively. n = 6 per group, *p < 0.05, **p < 0.01, ***p < 0.001 compared with control, determined by two-way ANOVA. e In vitro H-2-deoxy-glucose uptake by epididymal WAT was measured in the presence or absence of human insulin (10 mU/ml) (left panels). n = 6 per group, **p < 0.01, ***p < 0.001 compared with the insulin-treated control, determined by two-way ANOVA. IRS-PI3K signaling was evaluated by immunoblot analysis (right panels) after in vivo insulin stimulation by injection into the inferior vena cava (n = 2 per group). f Relative mRNA expression levels of hepatic gluconeogenic genes. n = 6 per group, *p < 0.05, **p < 0.01 compared with control analyzed mechanistic target of rapamycin complex 1 worsens the hyperlipidemic state in HFD-fed mice by (mTORC1) signaling in HFD-fed mice infected with the exerting opposing effects. indicated adenoviruses because mTORC1 signaling has previously been shown to be reciprocally regulated with Discussion respect to ketogenesis . In contrast to the effects of Ad- Growing evidence underscores the importance of Smad7 infection, liver lysates from HFD-fed mice Smads in obesity-associated metabolism, although Smad infected with Ad-Smads2/3/4 displayed lower levels of proteins are best known for their roles as transcription 240/244 phospho-S6 Ser in response to fasting compared factors in the TGF‐β signaling pathway. However, the with the levels in uninfected HFD-fed mice (Fig. 6h, left mechanisms underlying the regulation of cellular meta- panels). Immunoblot analyses of the mTORC1 signaling bolism by Smads remain poorly understood. We thus pathway also confirmed that Smads2/3/4 and Smad7 are investigated the role of Smads in the regulation of the responsible for downregulating and upregulating activity of the protein kinase MPK38, which plays a cri- mTORC1 signaling, respectively (Fig. 6h, right panels). tical role in the coordinate activation of ASK1/TGF‐β/ These results suggest that Smads2/3/4 can significantly p53 signaling that is closely associated with metabolic 25,26,30,31 improve the hyperlipidemic phenotype in HFD-fed mice homeostasis , and found that Smad proteins dif- by upregulating MPK38-dependent ASK1/TGF-β/ ferentially regulate MPK38 function through direct p53 signaling (Supplementary Fig. 4) and down- interactions, suggesting that Smad proteins, in addition to regulating mTORC1 signaling (Fig. 6h), and that Smad7 their roles as transcription factors, may function as Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 10 of 13 Fig. 6 Specific regulation of lipid metabolism and inflammation by adenoviral delivery of Smads to diet-induced obese mice. a-c Relative mRNA expression levels of lipogenic genes in epididymal WAT, and blood free fatty acid concentration a, lipogenic capacity of hepatocytes a, liver triglyceride concentration a, circulating total cholesterol, HDL-C, and LDL-C concentrations b, and serum concentrations of proinflammatory proteins c. n = 6 per group, *p< 0.05, **p < 0.01, ***p < 0.001 compared with control. d Measurement of β-oxidation using C-labeled palmitate in liver. n = 6 per group, *p < 0.05, **p < 0.01 compared with control. e Relative mRNA expression levels of fatty acid oxidative genes in epididymal WAT (left) and the isoproterenol-stimulated lipolytic response in isolated adipocytes (right). n = 6 per group, *p < 0.05, **p < 0.01, ***p < 0.001 compared with control. f Representative images of H&E-stained sections of livers. n = 6 per group. Scale bar, 100 μm. g Measurement of total ketone bodies in fed and fasted (24 h) blood. n = 6 per group, *p < 0.05, **p < 0.01 compared with fasted controls, determined by two-way ANOVA. h Phospho-S6 240/244 Ser levels in liver lysates from ad libitum-fed, fasted (24 h), and re-fed (2 h) HFD-fed mice infected or not (control) with the indicated adenoviruses (left panels). Immunoblot analyses of the mTORC1 signaling pathway using liver lysates (right panels). n = 6 per group regulators of MPK38. To investigate the mechanism biochemical analyses (Figs. 1 and 2). Smads2/3/4 mark- whereby MPK38 activity is regulated by Smads, we edly stimulate ASK1/TGF-β/p53 signaling, whereas examined whether the MPK38-mediated phosphorylation Smad7 inhibits this. However, such effects were not of Smads could affect the activity of MPK38, using various observed in the presence of MPK38-mediated Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 11 of 13 phosphorylation-defective Smad mutants. These findings In conclusion, our data show that Smad proteins differ- indicate that Smad proteins have specific regulatory entially regulate glucose and lipid metabolism and inflam- effects on MPK38 activity, depending on their class, but mation in diet-induced obese mice by differentially also that these effects are strictly dependent on Smad regulating MPK38‐dependent ASK1/TGF-β/p53 signaling, phosphorylation by MPK38. All these results strongly and that this effect is dependent on MPK38-mediated Smad suggest that Smad proteins have novel roles in the dif- phosphorylation and Smad class. Smads2/3/4 function as positive regulators of MPK38, whereas Smad7 functions as ferential regulation of cellular protein kinases. In addition to the effects on MPK38, we also observed a similar trend a negative regulator of MPK38. Moreover, this novel in the effect of Smads on ASK1 activity . function of Smads as differential regulators of MPK38‐ To further enhance understanding of the mechanism(s) dependent ASK1/TGF‐β/p53 signaling provides important by which Smad proteins differentially regulate MPK38 information about the mechanism determining how each activity, we also found that they modify the stability and class of Smad protein contributes to the maintenance of ubiquitination of MPK38. However, such effects were not metabolic homeostasis in mice. detected in MPK38-mediated phosphorylation-defective Smad mutants, indicating that Smad phosphorylation by Materials and methods MPK38 is required for these regulatory effects (Fig. 3). Antibodies, plasmids, chemicals, cell culture, and isolation These results suggest that Smad proteins differentially of hepatocytes and adipocytes regulate complex formation between MPK38 and its known Antibodies and plasmids used for experiments have 7,8,23 regulators, Trx and ZPR9, through direct interactions with been described previously . Cycloheximide (CHX) was MPK38, because Trx and ZPR9 destabilize and stabilize from Sigma-Aldrich. All other chemicals used was 76 252 8,23 MPK38 through Thr and Thr phosphorylation, described . Cell culture and isolation of hepatocytes and 23,24 8,23 respectively . Indeed, the present study demonstrates adipocytes were also described . that Smads2/3/4 decrease MPK38-Trx complex formation and increase MPK38-ZPR9 complex formation, whereas Generation of Smad KI cell lines Smad7 has the opposite effect (Fig. 4). These observations Genomic mutations were generated in HEK293 cells support a model in which Smad proteins differentially using the CRISPR/Cas9 system, as described previously . modify MPK38-Trx and/or MPK38-ZPR9 complex forma- Briefly, single-guide (sg) RNAs were designed to target the tion depending on their class and contribute to the differ- genomic areas adjacent to the Smad mutation sites (Sup- ential regulation of MPK38 kinase activity, resulting in plementary Table 1). Two complementary oligonucleotides specific regulation of MPK38-dependent ASK1/TGF‐β/ (Supplementary Table 2) containing the appropriate Smad p53 signaling pathways (Supplementary Fig. 5). guide sequence and Bbs1 ligation adapters were synthesized Obese mice display lower MPK38 kinase activity , by Bioneer Ltd. (Cheongwon, Korea). The annealed oligo- ASK1/TGF-β/p53 signaling , and Smads2/3/4 expression, nucleotides were ligated into a Bbs1-digested pX458 vector and higher levels of Smad7 expression (Supplementary (Addgene plasmid no. 48138) using the Quick-Ligation Fig. 6), compared with control mice. These results raise system (New England BioLabs). To generate the Smad KI the possibility that Smad proteins are involved in obesity- cell lines, HEK293 cells were cultured on a 24-well plate to associated metabolism, probably by regulating the acti- ~60% confluence and co-transfected with 1 μgSmadsg vation of ASK1/TGF‐β/p53 signaling through MPK38. To RNA plasmid and pUC19 Smad (Smad2 S245A, Smad3 test this hypothesis, we employed an adenoviral delivery S204A, Smad4 S343A, or Smad7 T96A) using Lipofecta- system to restore the lower levels of ASK1/TGF-β/ mine 2000 (Invitrogen). After culturing in a 96-well plate, p53 signaling in obese mice and analyzed the effect of GFP-positive cells were identified, followed by genomic Smads on obesity-associated metabolic abnormalities in DNA extraction. Smads were amplified by PCR using HFD-fed mice. Forced expression of Smads2/3/4, but not Smad-specific PCR primer pairs (Supplementary Table 3). MPK38-mediated phosphorylation-defective Smad The PCR products were A-tailed and cloned into the mutants, induced a significant increase in ASK1/TGF-β/ pGEM-T Easy vector (Promega) to confirm the identity of p53 signaling activation (Supplementary Fig. 4) and individual Smad KI clones by DNA sequencing. The in vivo ameliorated glucose and lipid metabolism (Figs. 5 and 6) phosphorylation of Smads by MPK38 was validated using in HFD-fed mice versus uninfected mice fed a HFD, in vitro kinase assays (Supplementary Fig. 7). whereas Ad-Smad7 infection decreased ASK1/TGF-β/ p53 signaling and further worsened the impaired glucose Coimmunoprecipitation, immunoblotting, MPK38 kinase and lipid metabolism in a MPK38 phosphorylation- assay, apoptosis assay, RNA isolation, and quantitative PCR dependent manner. In addition, MPK38 kinase activity (qPCR) was increased by Smads2/3/4 but decreased by Smad7 in Coimmunoprecipitation, immunoblot analysis, and HFD-fed mice (Supplementary Fig. 4). MPK38 kinase assay were performed as previously Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 12 of 13 described using Smad KI cells or HEK293 cells transiently 10 plaque-forming units) were directly injected into the transfected with the indicated expression vectors or tail vein or epididymal fat pads of 12- to 14-week-old 7,11 Smad-specific siRNAs . Apoptosis and qPCR were HFD-fed mice. 8,11 performed as described previously . Data analysis Glucose and insulin tolerance tests (GTTs/ITTs), Data are expressed as mean ± standard error and are lipogenesis/lipolysis assays, and blood metabolic representative of at least three independent experiments. parameters Statistical significance was determined by one-way or Plasma glucose levels for GTTs/ITTs were measured two-way analysis of variance (ANOVA), followed by using blood obtained from the tail vein at 0, 10, 20, 30, 45, Tukey’s multiple comparison test, using GraphPad Prism 60, 90, and 120 min post-injection according to the software (GraphPad Software). Mouse Metabolic Phenotyping Center recommendations 32 Acknowledgements for data presentation . Lipogenesis and lipolysis were This work was supported by a National Research Foundation of Korea Grant assayed in hepatocytes and adipocytes obtained from (2015R1A2A2A01006098) to H.H. and in part by a National Research Foundation of Korea Grant (2012R1A1A3015539) to H.-A.S. HFD-fed mice infected with the indicated adenoviruses, as described previously . Serum was obtained by cen- Author details trifugation of blood samples collected from the abdominal 1 Department of Biochemistry, School of Biological Sciences, Chungbuk aorta of 12–14-week-old HFD-fed mice and stored at National University, Cheongju 28644, Republic of Korea. Department of Biochemistry, University of Madras, Guindy Campus, Chennai 600025, India −70 °C. Serum levels of tumor necrosis factor (TNF)-α, IL-1β, IL-6, and monocyte chemoattractant protein 1 Authors' contributions (MCP1) were determined using analysis kits from H.-A.S. and R.M. conducted experiments and analyzed data. H.H. designed experiments and wrote the manuscript. Peprotech (ADI-900-047), RayBio (ELM-IL1beta-001), Invitrogen (KMC0061), and R&D Systems (MJE00), Conflict of interest respectively. Other blood metabolic parameters were The authors declare that they have no conflict of interest. measured as described previously . Publisher's note Serum insulin and glucose, liver triglyceride, 2-deoxy- Springer Nature remains neutral with regard to jurisdictional claims in glucose uptake, total ketone, and fatty acid β-oxidation published maps and institutional affiliations. Serum insulin and glucose levels, liver triglyceride Supplementary Information accompanies this paper at https://doi.org/ content, and 2-deoxy-glucose uptake in epididymal WAT 10.1038/s41419-018-0489-x. and soleus muscles were measured as described pre- 8,33 viously . Total ketone levels were measured using a Received: 3 October 2017 Revised: 15 March 2018 Accepted: 19 March colorimetric assay from Wako Chemicals . Fatty acid oxidation rate was determined using [1- C] palmitic acid, as described previously . Animal experiments and adenoviral infection References Four-week-old male C57BL/6 mice purchased from 1. Wrighton, K. H., Lin, X. & Feng, X. H. Phospho-control of TGF-beta superfamily Orient (Seongnam, Korea) were fed a HFD (60% kcal as signaling. Cell Res. 19,8–20 (2009). 2. Bierie, B. & Moses, H. L. Tumour microenvironment: TGFbeta: the molecular fat diet, D12492; Research Diets, Inc.) for 8–10 weeks Jekyll and Hyde of cancer. Nat. Rev. Cancer 6,506–520 (2006). under a 12:12-h light:dark cycle. Male 12- to 14-week-old 3. Massagué, J., Blain, S. W. & Lo, R. S. TGFbeta signaling in growth control, cancer, and heritable disorders. Cell 103,295–309 (2000). C57BL/6 mice fed with a standard rodent diet were 4. Siegel,P.M., Shu, W.,Cardiff, R.D., Muller, W. J. &Massagué, J. Transforming used as controls. All procedures were performed in a growth factor beta signaling impairs Neu-induced mammary tumorigenesis specific pathogen-free facility at Chungbuk National while promoting pulmonary metastasis. Proc. Natl. Acad. Sci. USA 100, 8430–8435 (2003). University, in accordance with the approved animal pro- 5. Massagué,J.TGFbeta in. Cancer Cell. 134,215–230 (2008). tocols and the guidelines established by the Ethics Review 6. Li, M. O. & Flavell, R. A. TGF-beta: a master of all T cell trades. Cell 134,392–404 Committee of Chungbuk National University for Animal (2008). 7. Seong, H. A., Jung, H. & Ha, H. Murine protein serine/threonine kinase 38 sti- Experiments (CBNUA-966-16-02). Recombinant adeno- mulates TGF-beta signaling in a kinase-dependent manner via direct phos- viruses were generated as previously described .To phorylation of Smad proteins. J. Biol. Chem. 285, 30959–30970 (2010). prepare adenoviruses expressing wild-type and MPK38 8. Seong, H. A.,Manoharan,R.& Ha, H. Coordinateactivation ofredox- dependent ASK1/TGF-β signaling by a multiprotein complex (MPK38, ASK1, phosphorylation-defective Smad mutants, FLAG-tagged SMADs, ZPR9, and TRX) improves glucose and lipid metabolism in mice. wild-type and mutant Smad plasmids were used as tem- Antioxid. Redox Signal. 24,434–452 (2016). plates for PCR with the primers shown in Supplementary 9. Saito, R., Tabata, Y., Muto, A., Arai, K. & Watanabe, S. Melk-like kinase plays a role Table 4. Recombinant adenoviruses (approximately 1 × in hematopoiesis in the zebra fish. Mol. Cell. Biol. 25,6682–6693 (2005). Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 13 of 13 10. Beullens, M. et al. Substrate specificity and activity regulation of protein kinase 23. Seong, H. A., Manoharan, R. & Ha, H. Zinc finger protein ZPR9 functions as an MELK. J. Biol. Chem. 280, 40003–40011 (2005). activator of AMPK-related serine/threonine kinase MPK38/MELK involved in 11. Jung, H., Seong, H. A. & Ha, H. Murine protein serine/threonine kinase 38 ASK1/TGF-β/p53 signaling pathways. Sci. Rep. 7, 42502 (2017). activates apoptosis signal-regulating kinase 1 via Thr 838 phosphorylation. 24. Manoharan, R., Seong, H. A. & Ha, H. Thioredoxin inhibits MPK38-induced J. Biol. Chem. 283, 34541–34553 (2008). ASK1, TGF‐β, and p53 function in a phosphorylation-dependent manner. Free. 12. Gray, D. et al. Maternal embryonic leucine zipper kinase/murine protein serine- Radic. Biol. Med. 63,313–324 (2013). threonine kinase 38 is a promising therapeutic target for multiple cancers. 25. Tan,C.K., Chong, H. C.,Tan,E.H.&Tan, N. S. Getting ‘Smad’ about obesity and Cancer Res. 65, 9751–9761 (2005). diabetes. Nutr. Diabetes 2, e29 (2012). 13. Vulsteke, V. et al. Inhibition of spliceosome assembly by the cell cycle- 26. Berkers,C.R., Maddocks,O. D., Cheung,E. C., Mor, I. & Vousden,K. H. Metabolic regulated protein kinase MELK and involvement of splicing factor NIPP1. J. Biol. regulation by p53 family members. Cell. Metab. 18,617–633 (2013). Chem. 279,8642–8647 (2004). 27. van Diepen, J. A., Berbée, J. F., Havekes, L. M. & Rensen, P. C. Interactions 14. Higaki, M. & Shimokado, K. Phosphatidylinositol 3-kinase is required for growth between inflammation and lipid metabolism: relevance for efficacy of anti- factor-induced amino acid uptake by vascular smooth muscle cells. Arter. inflammatory drugs in the treatment of atherosclerosis. Atherosclerosis 228, Thromb. Vasc. Biol. 19,2127–2132 (1999). 306–315 (2013). 15. Seong, H. A., Jung, H., Kim, K. T. & Ha, H. 3-Phosphoinositide-dependent PDK1 28. Muoio,D.M. & Newgard, C. B. Molecular and metabolic mechanisms of insulin negatively regulates transforming growth factor-beta-induced signaling in a resistance and β-cell failure in type 2 diabetes. Nat. Rev. Mol. Cell. Biol. 9, kinase-dependent manner through physical interaction with Smad proteins. 193–205 (2008). J. Biol. Chem. 282, 12272–12289 (2007). 29. Sengupta, S., Peterson, T. R., Laplante, M., Oh, S. & Sabatini, D. M. mTORC1 16. Tan,C.K. etal. Smad3deficiency in mice protects against insulin resistance controls fasting-induced ketogenesis and its modulation by ageing. Nature and obesity induced by a high-fat diet. Diabetes 60,464–476 (2011). 468, 1100–1104 (2010). 17. Yadav, H. et al. Protection from obesity and diabetes by blockade of TGF-β/ 30. Blüher, M. et al. Activated Ask1-MKK4-p38MAPK/JNK stress signaling pathway Smad3 signaling. Cell. Metab. 14,67–79 (2011). in human omental fat tissue may link macrophage infiltration to whole-body 18. Tsurutani, Y. et al. The roles of transforming growth factor-β and Insulin sensitivity. J. Clin. Endocrinol. Metab. 94, 2507–2515 (2009). Smad3 signaling in adipocyte differentiation and obesity. Biochem. Biophys. 31. Hattori, K. et al. ASK1 signalling regulates brown and beige adipocyte function. Res. Commun. 407,68–73 (2011). Nat. Commun. 7, 11158 (2016). 19. Lin, H. M. et al. Transforming growth factor-beta/Smad3 signaling regulates 32. McGuinness, O. P.,Ayala,J.E., Laughlin,M.R.&Wasserman, D. H. NIH insulin gene transcription and pancreatic islet beta-cell function. J. Biol. Chem. experiment in centralized mouse phenotyping: the Vanderbilt experience 284, 12246–12257 (2009). and recommendations for evaluating glucose homeostasis in the mouse. 20. Perry, J. R. et al. Interrogating type 2 diabetes genome-wide association data Am.J.Physiol.Endocrinol. Metab. 297, E849–E855 (2009). using a biological pathway-based approach. Diabetes 58,1463–1467 (2009). 33. Chang, P. Y. et al. Overexpression of hexokinase II in transgenic mice: evidence 21. Nomura, M. et al. SMAD2 disruption in mouse pancreatic beta cells leads to that increased phosphorylation augments muscle glucose uptake. J. Biol. islet hyperplasia and impaired insulin secretion due to the attenuation of ATP- Chem. 271,14834–14839 (1996). sensitive K+ channel activity. Diabetologia 57, 157–166 (2014). 34. Seong, H. A., Manoharan, R. & Ha, H. A crucial role for the phosphorylation of 22. Li, H. Y. et al. Amelioration of high fat diet-induced glucose intolerance by STRAP at Ser(188) by MPK38 in STRAP-dependent cell death through ASK1, blockade of Smad4 in pancreatic beta-cells. Exp. Clin. Endocrinol. Diabetes 123, TGF-β, p53, and PI3K/PDK1 signaling pathways. Cell Cycle 13,3357–3374 221–226 (2015). (2014). Official journal of the Cell Death Differentiation Association http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Death & Disease Springer Journals

Smad proteins differentially regulate obesity-induced glucose and lipid abnormalities and inflammation via class-specific control of AMPK-related kinase MPK38/MELK activity

Free
13 pages

Loading next page...
 
/lp/springer_journal/smad-proteins-differentially-regulate-obesity-induced-glucose-and-eJc3emJpz0
Publisher
Nature Publishing Group UK
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Life Sciences, general; Biochemistry, general; Cell Biology; Immunology; Cell Culture; Antibodies
eISSN
2041-4889
D.O.I.
10.1038/s41419-018-0489-x
Publisher site
See Article on Publisher Site

Abstract

Smad proteins have been implicated in metabolic processes, but little is known about how they regulate metabolism. Because Smad 2, 3, 4, and 7 have previously been shown to interact with murine protein serine–threonine kinase 38 (MPK38), an AMP‐activated protein kinase (AMPK)-related kinase that has been implicated in obesity-associated metabolic defects, we investigated whether Smad proteins regulate metabolic processes via MPK38. Smads2/3/4 increased, but Smad7 decreased, MPK38-mediated apoptosis signal-regulating kinase-1 (ASK1)/transforming growth factor-β (TGF-β)/p53 signaling. However, MPK38-mediated phosphorylation-defective Smad mutants (Smad2 S245A, Smad3 S204A, Smad4 S343A, and Smad7 T96A) had no such effect. In addition, Smads2/3/4 increased, but Smad7 decreased, the stability of MPK38. Consistent with this, Smads2/3/4 attenuated complex formation between MPK38 and its negative regulator thioredoxin (Trx), whereas Smad7 increased this complex formation. However, an opposite effect was observed on complex formation between MPK38 and its positive regulator zinc-finger-like protein 9 (ZPR9). When Smads were overexpressed in high-fat diet (HFD)-fed obese mice using an adenoviral delivery system, Smads2/ 3/4 improved, but Smad7 worsened, obesity-associated metabolic parameters and inflammation in a MPK38 phosphorylation-dependent manner. These findings suggest that Smad proteins have class-specific impacts on obesity-associated metabolism by differentially regulating MPK38 activity in diet-induced obese mice. Introduction cell proliferation, differentiation, apoptosis, migration, The identification of a growing list of intracellular extracellular matrix remodeling, immune functions, and kinases that phosphorylate Smad proteins suggests that tumor metastasis. This occurs through the combined use the transforming growth factor-β (TGF-β)/Smad signaling of TGF-β signaling pathway components, such as Smads pathway cross-talks with a variety of other intracellular and Smad-interacting transcription factors, cross-talk signaling pathways . The TGF-β signaling pathway reg- with other intracellular signaling pathways, and the ulates a broad range of cellular processes, which include ability of TGF-β receptors to activate other 2–6 signaling modules . Many studies have shown that Smads are phosphorylated by multiple intracellular kina- ses, including mitogen-activated protein kinases, Correspondence: Hyunjung Ha (hyunha@cbnu.ac.kr) 1 2+ Department of Biochemistry, School of Biological Sciences, Chungbuk Ca /calmodulin-dependent kinase II, cyclin-dependent National University, Cheongju 28644, Republic of Korea kinase (CDK), protein kinase C, G protein-coupled Department of Biochemistry, University of Madras, Guindy Campus, Chennai receptor kinase 2, extracellular signal-regulated kinase, 600025, India Edited by C. Munoz-Pinedo © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; Seong et al. Cell Death and Disease (2018) 9:471 Page 2 of 13 apoptosis signal-regulating kinase-1 (ASK1), and murine molecular mechanisms involved in the regulation of meta- protein serine–threonine kinase 38 (MPK38)/maternal bolic homeostasis by TGF-β signaling remain poorly 1,7,8 embryonic leucine zipper kinase (MELK) , suggesting understood. that the TGF-β pathway is closely integrated with other In this study, we show that there are direct physical and intracellular signaling pathways to achieve tightly regu- functional interactions between MPK38 and Smads lated TGF-β responses. However, most of these studies (Smad2, 3, 4, and 7). Smads2/3/4 stimulate MPK38- dependent ASK1/TGF-β/p53 signaling pathways, whereas have focused on the regulatory role of Smad phosphor- ylation in the TGF-β signaling pathway. Additional stu- Smad7 inhibits these signaling pathways through differ- dies are required to investigate the effect of Smad proteins ential regulation of MPK38 activity. Furthermore, over- on the activity of these interacting kinases in order to expression of Smads2/3/4 improves, whereas Smad7 decipher the molecular interplay between TGF-β and overexpression worsens, obesity-associated metabolic other intracellular signaling pathways. parameters by differentially regulating MPK38 activity in MPK38/MELK, an AMP‐activated protein kinase HFD-induced obese mice. (AMPK)-related kinase, has been shown to mediate var- ious cellular functions, including proliferation, spliceo- Results some assembly, gene expression, carcinogenesis, MPK38 kinase activity is increased by Smads2/3/4 but 9–13 apoptosis, and metabolism , although its exact phy- decreased by Smad7 siological functions still remain to be determined. MPK38 Given that MPK38 interacts with and phosphorylates and its interacting partner Smad3 have recently been Smad proteins, leading to the activation of TGF-β signal- shown to serve as components of a multi-protein complex ing , we reasoned that Smad proteins would affect MPK38 linking ASK1 and TGF-β signaling pathways, which are activity through direct interaction and phosphorylation. To involved in glucose and lipid metabolism in mice, and to assess this possibility, we performed immunoblot analysis contribute to the activation of ASK1 signaling via a direct using CRISPR/Cas9-mediated Smad knock-in (KI) HEK293 8,11 interaction with ASK1 . TGF-β1 was previously repor- cells (Smad2 S245A, Smad3 S204A, Smad4 S343A, and ted to positively regulate the 3-phosphoinositide- Smad7 T96A), which are defective in MPK38-mediated 14 7 dependent protein kinase-1 (PDK1)/AKT1 pathway , phosphorylation , to determine the endogenous kinase although PDK1 was shown to inhibit TGF-β signaling activity of MPK38 in the presence or absence of ASK1/ through direct interactions with Smads . These findings TGF-β/p53 signals, including H O,TGF-β1, and 2 2 suggest potential roles of Smads in the regulation of key 5-fluorouracil (5FU). The endogenous kinase activity of kinases involved in intracellular signaling pathways that MPK38 was markedly lower in the S245A, S204A, and are integrated with TGF-β signaling. S343A KI cells when compared with wild-type control cells, Recent discoveries have shed some light on the important whereas T96A KI cells had higher MPK38 kinase activity role that TGF-β signaling plays in adipose physiology and (Fig. 1a). These results were corroborated by inhibition of 16–18 metabolism . Smad3 deficiency in mice resulted in the kinase activity of MPK38 using a potent MPK38 inhi- improved glucose tolerance and insulin sensitivity, accom- bitor OTSSP167 (Fig. 1a) . We also analyzed the effects of panied by reduced white adipose tissue (WAT) mass and phosphorylation of Smad isoforms by MPK38 in the reg- browning. The associated increase in mitochondrial bio- ulation of MPK38 kinase activity using in vitro kinase assays genesis resulted in the dissipation of the excess energy with recombinant wild-type and MPK38-mediated phos- 16,17 stored in WAT by thermogenesis .Higher TGF-β1in phorylation-defective Smad mutants. The MPK38 kinase humans has been shown to positively correlate with greater activity was increased by recombinant wild-type Smads2/3/ adiposity and a poor metabolic profile, and to negatively 4, but decreased by recombinant wild-type Smad7. How- correlate with fitness . Several recent studies have ever, the recombinant MPK38-mediated phosphorylation- demonstrated that TGF-β signaling regulates insulin gene defective Smad mutants had no such effects (Fig. 1b). These transcription in pancreatic β cells . Moreover, the Smad3 findings indicate that Smad proteins differentially regulate gene was identified in a genome-wide association study for the kinase activity of MPK38, resulting in specificSmad- type 2 diabetes risk .These findings implicate Smad3 as a mediated regulation of coordinate MPK38-induced ASK1/ potential target for the treatment of obesity and its asso- TGF-β/p53 signaling. ciated disorders. Conversely, targeted disruption of Smad2 in mouse pancreatic β cells caused islet cell hyperplasia and MPK38-dependent ASK1/TGF-β/p53-mediated impaired insulin secretion by attenuating ATP-sensitive K+ transcription, apoptosis, and signaling activation are channel activity . However, inhibition of Smad4 in pan- increased by Smads2/3/4 but decreased by Smad7 creatic β cells conferred minor but significant improve- Because MPK38 has been shown to potentiate coordi- 7,11,23 ments in blood glucose and glucose tolerance in high-fat nate ASK1/TGF-β/p53 signaling ,we first examined diet (HFD)-induced obese mice . Nevertheless, the the effect of Smad proteins on ASK1/TGF-β/p53- Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 3 of 13 Fig. 1 (See legend on next page.) Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 4 of 13 (see figure on previous page) Fig. 1 Differential regulation of MPK38 kinase activity by Smad proteins. a To assess the effects of wild-type (WT) and MPK38-mediated phosphorylation-defective Smad mutants (Smad2 S245A, Smad3 S204A, Smad4 S343A, and Smad7 T96A) on MPK38 kinase activity, WT and CRISPR/ Cas9 Smad knock-in HEK293 cells (S245A, S204A, S343A, and T96A) were incubated with (+) or without (−) OTSSP167 (1 μM, 2 h), a potent MPK38 inhibitor, and then treated with or without the following stimuli: H O (2 mM, 30 min), TGF‐β1 (2 ng/ml, 20 h), or 5FU (0.38 mM, 30 h). The cell lysates 2 2 were subjected to immunoprecipitation with antibodies for ZPR9, STRAP, and Trx, followed by immunoblot analysis using anti-phospho-specific 23,24,34 252 188 76 antibodies for ZPR9 Thr , STRAP Ser , and Trx Thr . The protein levels of MPK38 and Smads in cell lysates were examined with anti-MPK38 and anti-Smads antibodies. b Recombinant MPK38 proteins (~5 μg) were screened in in vitro kinase assays using recombinant ZPR9 (~5 μg) as a substrate in the presence of recombinant WT and MPK38-mediated phosphorylation-defective Smad mutants (~0.8 μg each). Ratios of P-ZPR9/ZPR9 32 32 a, P-ZPR9/ZPR9 b, and P-MPK38/MPK38 b were determined by density analysis of the bands, and the fold-increase relative to the untreated WT 32 32 controls a or control lacking Smad expression b is presented for each protein. P, P incorporation; P, phosphorylated; re., recombinant mediated transcription induced by MPK38. As shown in Smads (Smad2, 3, 4, and 7) on MPK38 protein stability. Fig. 2 and Supplementary Fig. 1, the ASK1/TGF-β/p53- HEK293 cells were transfected with expression vectors mediated transactivation induced by MPK38 increased in encoding wild-type and MPK38-mediated phosphoryla- a kinase-dependent manner in the presence of wild-type tion-defective Smad mutants, and MPK38 protein levels Smads2/3/4, whereas MPK38-mediated phosphorylation- were quantified using immunoblot analysis. As shown in defective Smad mutants had no such effect. By contrast, Fig. 3a, Smads2/3/4 increased the stability of MPK38 Smad7 decreased the transcriptional activity. These compared with control cells expressing empty vector, results suggest that Smads2/3/4 are positive regulators of whereas Smad7 decreased MPK38 stability. However, MPK38 and that Smad7 is a negative regulator of MPK38. such effects were not observed in the presence of MPK38- The data also demonstrate that the phosphorylation of mediated phosphorylation-defective Smad mutants. Smad proteins by MPK38 is critical for specific Smad- Conversely, treatment of Smad-expressing HEK293 cells mediated regulation of ASK1/TGF-β/p53-mediated with both cycloheximide and MG132, a proteasomal transactivation induced by MPK38. inhibitor, led to greater stability of MPK38, compared We next examined whether Smad proteins can influence with that in non-MG132-treated Smad-expressing ASK1/TGF-β/p53-mediated apoptosis induced by MPK38. HEK293 cells (Fig. 3a). These results indicate a critical As shown in Fig. 2 (middle panels) and Supplementary role for the proteasome pathway in the effects of Smads Fig. 1, the coexpression of wild-type Smads2/3/4 and on MPK38 degradation in cells. We then analyzed the MPK38 resulted in a kinase-dependent increase in ASK1/ effect of Smads on MPK38 ubiquitination using CRISPR/ TGF-β/p53-induced apoptosis compared with control cells Cas9-mediated Smad KI HEK293 cells treated with (+)or transfected with MPK38 alone. However, the stimulatory without (−) OTSSP167. Smads2/3/4 endogenously effect of Smads2/3/4 on ASK1/TGF-β/p53-induced apop- decreased the ubiquitination of MPK38 in a MPK38 tosis was not observed in the presence of MPK38-mediated phosphorylation-dependent manner, whereas Smad7 phosphorylation-defective Smad mutants. Conversely, the increased MPK38 ubiquitination (Fig. 3b). These results opposite effect was observed in the presence of Smad7. suggest that, in addition to the class of Smad protein, We also evaluated MPK38-mediated ASK1/TGF-β/ MPK38-mediated phosphorylation of Smad proteins also p53 signaling activation using CRISPR/Cas9-mediated plays a critical role in the regulation of MPK38 protein Smad KI HEK293 cells in the presence or absence of H O / stability. 2 2 TGF-β1/p53. As shown in Fig. 2 (right panels), and as We then investigated whether the specific Smad- expected, the activation of ASK1/TGF-β/p53 signaling was mediated regulation of MPK38 stability is dependent on markedly lower in S245A, S204A, and S343A KI cells when the interaction between MPK38 and Mdm2 . HEK293 compared with wild-type control cells, whereas T96A KI cells were transfected with expression vectors for wild- cells showed greater activation of ASK1/TGF-β/ type and MPK38-mediated phosphorylation-defective p53 signaling. Together, these findings suggest that Smad mutants. Smads2/3/4 markedly decreased MPK38- MPK38-induced ASK1/TGF-β/p53 activity is positively Mdm2 complex formation in a MPK38 phosphorylation- regulated by Smads2/3/4, but negatively regulated by dependent manner, whereas Smad7 increased complex Smad7, in a MPK38 phosphorylation-dependent manner. formation (Fig. 3c). These results also indicate the important role of MPK38-mediated phosphorylation of MPK38 stability is increased by Smads2/3/4 but decreased specific Smads in the differential regulation of Mdm2- by Smad7 dependent MPK38 ubiquitination. All these results sug- To address the mechanism of the differential regulation gest critical roles for specific Smad protein classes and of MPK38 activity by Smads, we investigated the effect of their phosphorylation by MPK38 in the regulation of Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 5 of 13 Fig. 2 (See legend on next page.) MPK38 stability: Smads2/3/4 stabilize MPK38, whereas siRNAs were subjected to immunoprecipitation using an Smad7 reduces its stability. anti-MPK38 antibody, followed by immunoblot analysis using an anti-Trx antibody. There was a dose-dependent Smad proteins differentially regulate complex formation decrease in endogenous complex formation between between MPK38 and its regulators, thioredoxin (Trx) and MPK38 and Trx in cells expressing Smads2/3/4 compared zinc-finger-like protein 9 (ZPR9) with cells not expressing Smads. By contrast, Smad7 To further explore the mechanism of differential reg- transfection increased complex formation (Fig. 4a). These ulation of MPK38 stability by Smads, we investigated results were corroborated by Smad silencing experiments whether Smad proteins affect the interaction between using Smad-specific siRNAs (Fig. 4a). These findings MPK38 and Trx, which destabilizes it . HEK293 cells indicate that Smad class type contributes to the Smad- transfected with various quantities of plasmid vectors mediated differential modulation of MPK38-Trx complex encoding Smads2/3/4/7 or specific Smad-targeting formation. Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 6 of 13 (see figure on previous page) Fig. 2 Differential regulation of MPK38‐mediated ASK1/TGF-β/p53 signaling by Smads. a 293T cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg), WT and K40R MPK38 (0.8 μg), WT ASK1 (0.6 μg), c‐fos (0.6 μg), or the AP‐1- luciferase plasmid (0.2 μg), as indicated. After 48 h, the cells were harvested, and luciferase activity was measured using a luciferase assay system (Promega). The total DNA concentration was kept constant by supplementation with empty vector DNA. The values were adjusted relative to the expression levels of a co-transfected β-galactosidase reporter control (left panels). HEK293 cells were transfected with various concentrations of vectors encoding WT and mutant Smads3/7 (0.6 and 1.2 μg), or WT and K40R MPK38 (0.8 μg), as indicated, in the presence or absence of H O (1 mM, 2 2 9 h). Cells exposed only to H O were used as a positive control. Apoptotic cell death was determined using a GFP system (middle panels). WT and 2 2 CRISPR/Cas9 Smad knock-in HEK293 cells were treated with (+) or without (−)H O (2 mM, 30 min), and the cell lysates were subjected to 2 2 immunoprecipitation with antibodies for ASK1, MKK3, p38, and ATF2, followed by immunoblot analysis using anti‐phospho-specific antibodies for 845 838 189/207 180 182 71 ASK1 Thr (Thr in human), MKK3/6 Ser , p38 Thr /Tyr , and ATF2 Thr . The protein levels of MPK38 and Smads in cell lysates were examined with anti-MPK38 and anti-Smads antibodies (right panels). *p < 0.05, **p < 0.01, ***p < 0.001 compared with MPK38 alone in the presence of ASK1 or H O . b HaCaT cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg), WT and 2 2 K40R MPK38 (0.4 μg), or p3TP‐Lux plasmid (0.2 μg), as indicated, in the presence or absence of TGF‐β1 (100 pM) (left panels). HaCaT cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg) and/or WT and K40R MPK38 (0.4 μg), as indicated, together with an expression vector encoding GFP (1 μg). After treatment of the transfected cells with TGF‐β1 (2 ng/ml, 20 h), apoptotic cell death was then determined (middle panels). WT and CRISPR/Cas9 Smad knock-in HEK293 cells were treated with (+) or without (−) TGF‐β1 (2 ng/ml, 20 h), and the cell lysates were analyzed by immunoblot analysis using antibodies for PAI-1, p21, Smad7, CDK4, cyclin D1, MPK38, and Smads (Smad2, Smad3, Smad4, and Smad7) (right panels). **p < 0.01, ***p < 0.001 compared with MPK38 alone in the presence of TGF-β1. c MCF7 cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg), WT and K40R MPK38 (0.4 μg), or p53-Luc plasmid (0.2 μg), as indicated, in the presence or absence of p53 (0.3 μg) (left panels). MCF7 cells were transfected with various concentrations of vectors encoding WT or mutant Smads3/7 (0.5 and 1 μg) and/or WT and K40R MPK38 (0.4 μg), as indicated, together with an expression vector encoding GFP (1 μg) in the presence or absence of p53 (0.6 μg). Apoptotic cell death was then determined (middle panels). WT and CRISPR/Cas9 Smad knock-in HEK293 cells were treated with (+) or without (−) 5FU (0.38 mM, 30 h), and the cell lysates were analyzed by immunoblot analysis using antibodies for p53, p21, Mdm2, Bax, MPK38, and Smads (Smad2, Smad3, Smad4, and Smad7) (right panels). ***p < 0.001 compared with MPK38 alone in the presence of p53. The results represent the mean ± S.E. of at least three independent experiments performed in duplicate. Kinase-dead MPK38, K40R Because ZPR9 was recently shown to function as a adipocytes, compared with uninfected HFD-fed mice or stabilizer of MPK38 , we next investigated the effect of HFD-fed mice infected only with Green Fluorescent Smads on ZPR9 binding to MPK38. As expected, Smads2/ Protein (GFP)-expressing adenovirus (Ad-GFP) (Fig. 5a; 3/4 increased endogenous complex formation between Supplementary Fig. 2a). The mRNA expression levels of MPK38 and ZPR9, but Smad7 decreased this (Fig. 4b). key adipogenic regulators, including CCAAT-enhancer- Consistent with this, endogenous Smad silencing using binding protein α, peroxisome proliferator-activated Smad-specific siRNAs had the opposite effect on MPK38- receptor γ (PPARγ), and fatty acid binding protein 4, ZPR9 complex formation (Fig. 4b). These observations were significantly lower in WAT from HFD-fed mice provide evidence that the differential regulation of infected with Ad-Smads2/3/4 than those in the uninfected MPK38 stability by Smads is mediated by modulating HFD-fed mice. However, Ad-Smad7 infection had an complex formation between MPK38 and its regulators, opposite effect (Fig. 5b). Trx and ZPR9. Fasting blood glucose was lower in HFD-fed mice infected with Ad-Smads2/3/4 than in uninfected HFD-fed Smads2/3/4 improve, but Smad7 worsens, glucose mice (Fig. 5c) and Ad-Smads2/3/4 infection enhanced metabolism in a mouse model of diet-induced obesity glucose tolerance and insulin sensitivity in HFD-fed mice Emerging evidence has implicated ASK1/TGF-β/ (Figs. 5c, d). In parallel, mice infected with Ad-Smads2/3/ p53 signaling pathways in the pathogenesis of obesity- 4 exhibited lower circulating insulin levels under fasting 8,25,26 associated metabolic diseases . Our recent studies conditions, whereas Ad-Smad7 infection further have shown that both genetically and diet-induced obese increased the levels of circulating insulin (Fig. 5d, left). In mice display lower levels of ASK1/TGF-β/p53 signaling addition, Ad-Smads2/3/4 infection significantly increased and Smad3 expression, and a higher level of Smad7 in vitro insulin-stimulated 2-deoxy-glucose uptake in expression, when compared with control wild-type or WAT and muscle, whereas Ad-Smad7 infection had the 8,23 chow-fed mice . Based on these findings, an adenoviral opposite effect (Fig. 5e, left; Supplementary Fig. 2b). delivery system was employed to investigate whether Consistent with this, we observed that Ad-Smads2/3/4 Smad proteins regulate obesity-associated glucose meta- infection caused a significant upregulation of the insulin bolism by differentially regulating ASK1/TGF-β/ receptor substrate (IRS)-phosphoinositide 3-kinase (PI3K) p53 signaling in diet-induced obese mice. The pathway, which may contribute to enhanced glucose adenoviral delivery of Smads2/3/4 significantly decreased uptake, whereas the IRS-PI3K pathway was down- cellular distribution toward extremely large, hypertrophic regulated by Ad-Smad7 infection (Fig. 5e, right; Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 7 of 13 Fig. 3 Differential regulation of MPK38 stability by Smads. a MPK38 protein stability was assessed by immunoblot analysis using an anti‐MPK38 antibody. HEK293 cells were transfected with pCMV2‐FLAG (vector) or vectors encoding FLAG-tagged WT or mutant Smads. Time intervals indicate the number of minutes after treatment with cycloheximide (CHX, 20 μg/ml) alone or with MG132 (10 μM). The MPK38 levels in at least three independent experiments were quantified by densitometry (right panels). The relative densitometry at each time point was expressed as a percentage of the density at time 0 min after normalization the corresponding β-actin level. b The ubiquitination of endogenous MPK38 was assessed using WT and CRISPR/Cas9 Smad knock-in HEK293 cells treated with (+) or without (−) OTSSP167 (1 μM, 2 h). c HEK293 cells were transfected with vectors encoding FLAG-tagged WT or mutant Smads, as indicated. Cell lysates were subjected to immunoprecipitation using an anti-MPK38 antibody (IP: α-MPK38) followed by immunoblot analysis using an anti‐Mdm2 antibody to determine the endogenous levels of MPK38‐ Mdm2 complexes. Ratio of Mdm2/MPK38 was determined by density analysis of the bands, and the fold-increase relative to control not expressing Smads is presented Supplementary Fig. 1b). HFD-fed mice infected with Ad- 3/4 infection significantly reduced the mRNA expression Smads2/3/4 displayed a considerable decrease in blood of adipose and hepatic lipogenic genes, including fatty glucose levels (Supplementary Fig. 2c), together with acid synthase (FAS), sterol CoA desaturase 1, and sterol lower mRNA expression levels of hepatic gluconeogenic regulatory element-binding transcription factor 1c, con- genes, including glucose-6-phosphatase (G6PC), phos- sistent with lower circulating free fatty acid levels, phoenolpyruvate carboxykinase-1 (PCK1), and peroxi- whereas Ad-Smad7 infection had the opposite effect in some proliferator-activated receptor γ coactivator 1α both tissues (Fig. 6a; Supplementary Fig. 3a). Consistent (PGC1α), when compared with uninfected HFD-fed mice with this, and in contrast to the effects of infection with (Fig. 5f). By contrast, no MPK38-mediated phosphoryla- Ad-Smad7, HFD-fed mice infected with Ad-Smads2/3/4 tion-defective Smad mutants had such an effect (Fig. 5; exhibited lower lipogenesis (Fig. 6a), liver triglyceride Supplementary Fig. 2). These results indicate that (Fig. 6a), circulating total cholesterol, high-density lipo- Smads2/3/4 improve, but Smad7 worsens, glucose meta- protein (HDL)-cholesterol, and low-density lipoprotein bolism in a MPK38 phosphorylation-dependent manner (LDL)-cholesterol (Fig. 6b). However, there was no such in diet-induced obese mice. effect in the presence of MPK38-mediated phosphoryla- tion-defective Smad mutants. These results suggest that Smads2/3/4 improve, but Smad7 worsens, lipid Smad phosphorylation by MPK38 is important in the metabolism and inflammation in a mouse model of diet- Smad-mediated differential regulation of lipogenesis. A induced obesity previous report demonstrated that cholesterol, fatty acids, We then examined whether Smad proteins regulate and modified lipids activate inflammatory pathways and lipogenic gene expression in HFD-fed mice. Ad-Smads2/ modulate the activity of leukocytes . Therefore, we also Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 8 of 13 Fig. 4 Smad-specific regulation of complex formation between MPK38 and Trx or ZPR9. HEK293 cells were transfected with various concentrations of vectors encoding GST-tagged Smads (0.5 and 1 μg) or Smad-specific siRNAs (100 and 200 nM), as indicated. Cell lysates were subjected to immunoprecipitation with an anti-MPK38 antibody (IP: α-MPK38), and endogenous complex formation between MPK38 and its regulators, Trx a and ZPR9 b, was assessed by immunoblot analysis using anti-Trx and anti-ZPR9 antibodies (upper panels). Ratios of Trx/MPK38 a and ZPR9/MPK38 b were determined by density analysis of the bands, and the fold-increase relative to controls not expressing Smads is presented (lower panels). Each experiment was repeated at least three times with similar results examined the expression of proinflammatory genes in (Fig. 6d). Indeed, WAT from HFD-fed mice infected with blood samples. Ad-Smads2/3/4 infection considerably Ad-Smads2/3/4 exhibited higher expression of key genes decreased serum proinflammatory proteins when com- involved in fatty acid oxidation, such as peroxisome pared with uninfected HFD-fed mice, whereas Ad-Smad7 proliferator-activated receptor α (PPARα), carnitine pal- infection increased the serum levels of proinflammatory mitoyltransferase 1, and acyl-CoA oxidase, as well as proteins (Fig. 6c). However, these effects were not lower levels of blood triglycerides and higher observed after infection with MPK38-mediated phos- isoproterenol-stimulated lipolysis (Fig. 6e; Supplementary phorylation-defective Smad mutants. These results indi- Fig. 3c). HFD-fed mice infected with Ad-Smads2/3/4 also cate that Smads2/3/4 have beneficial effects, whereas displayed substantial decreases in liver lipid accumulation Smad7 has an adverse effect, on lipogenesis and and triglyceride content (Fig. 6a, f). However, we observed inflammation. the opposite trend in HFD-fed mice infected with Ad- Ad-Smads2/3/4 infections also increased the mRNA Smad7. All these results indicate the positive roles of expression of adipose lipolytic genes, including hormone- Smads2/3/4 and the negative role of Smad7 in lipid sensitive lipase (HSL), adipose triglyceride lipase (ATGL), oxidation. and beta-3 adrenergic receptor (ADRB3), in a MPK38 Because abnormal mitochondrial fat oxidation is phosphorylation-dependent manner, whereas Ad-Smad7 associated with insulin resistance and impaired keto- infection decreased the expression of these genes (Sup- genesis , we also investigated whether Smad proteins plementary Fig. 3b). Consistent with this finding, Ad- differentially regulate hepatic ketogenesis. As expected, Smads2/3/4 infection stimulated hepatic fatty acid utili- Ad-Smads2/3/4 infection enhanced ketone body pro- zation by stimulating the oxidation of fatty acids via duction (Fig. 6g) and ketogenic gene expression (Sup- mitochondrial and peroxisomal β-oxidation pathways in a plementary Fig. 3d) under fasting conditions when MPK38 phosphorylation-dependent manner, whereas compared with uninfected HFD-fed mice, but Ad- Smad7 had the opposite effect on fatty acid utilization Smad7 infection had the opposite effect. We also Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 9 of 13 Fig. 5 Specific effects of adenoviral delivery of Smads on glucose metabolism in diet-induced obese mice. a Hematoxylin and eosin (H&E)- stained paraffin-embedded sections of epididymal WAT in HFD-fed mice infected with the indicated adenoviruses. Scale bar, 100 μm. b Relative mRNA expression levels of adipogenic genes. mRNA expression was quantified by densitometry, and the fold-increase relative to control is presented. n = 6 per group, *p < 0.05, **p < 0.01 compared with control. c, d Blood glucose c and insulin d levels in fed and fasting (16 h) HFD-fed mice that had been infected with the indicated adenoviruses or not (control) (left panels). n = 6 per group, *p < 0.05, **p < 0.01 compared with the fasted controls, determined by two-way ANOVA. Glucose tolerance tests c and insulin tolerance tests d were conducted by measuring blood glucose concentrations in mice following intraperitoneal injection of glucose (2 g/kg) or insulin (0.75 U/kg), respectively. n = 6 per group, *p < 0.05, **p < 0.01, ***p < 0.001 compared with control, determined by two-way ANOVA. e In vitro H-2-deoxy-glucose uptake by epididymal WAT was measured in the presence or absence of human insulin (10 mU/ml) (left panels). n = 6 per group, **p < 0.01, ***p < 0.001 compared with the insulin-treated control, determined by two-way ANOVA. IRS-PI3K signaling was evaluated by immunoblot analysis (right panels) after in vivo insulin stimulation by injection into the inferior vena cava (n = 2 per group). f Relative mRNA expression levels of hepatic gluconeogenic genes. n = 6 per group, *p < 0.05, **p < 0.01 compared with control analyzed mechanistic target of rapamycin complex 1 worsens the hyperlipidemic state in HFD-fed mice by (mTORC1) signaling in HFD-fed mice infected with the exerting opposing effects. indicated adenoviruses because mTORC1 signaling has previously been shown to be reciprocally regulated with Discussion respect to ketogenesis . In contrast to the effects of Ad- Growing evidence underscores the importance of Smad7 infection, liver lysates from HFD-fed mice Smads in obesity-associated metabolism, although Smad infected with Ad-Smads2/3/4 displayed lower levels of proteins are best known for their roles as transcription 240/244 phospho-S6 Ser in response to fasting compared factors in the TGF‐β signaling pathway. However, the with the levels in uninfected HFD-fed mice (Fig. 6h, left mechanisms underlying the regulation of cellular meta- panels). Immunoblot analyses of the mTORC1 signaling bolism by Smads remain poorly understood. We thus pathway also confirmed that Smads2/3/4 and Smad7 are investigated the role of Smads in the regulation of the responsible for downregulating and upregulating activity of the protein kinase MPK38, which plays a cri- mTORC1 signaling, respectively (Fig. 6h, right panels). tical role in the coordinate activation of ASK1/TGF‐β/ These results suggest that Smads2/3/4 can significantly p53 signaling that is closely associated with metabolic 25,26,30,31 improve the hyperlipidemic phenotype in HFD-fed mice homeostasis , and found that Smad proteins dif- by upregulating MPK38-dependent ASK1/TGF-β/ ferentially regulate MPK38 function through direct p53 signaling (Supplementary Fig. 4) and down- interactions, suggesting that Smad proteins, in addition to regulating mTORC1 signaling (Fig. 6h), and that Smad7 their roles as transcription factors, may function as Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 10 of 13 Fig. 6 Specific regulation of lipid metabolism and inflammation by adenoviral delivery of Smads to diet-induced obese mice. a-c Relative mRNA expression levels of lipogenic genes in epididymal WAT, and blood free fatty acid concentration a, lipogenic capacity of hepatocytes a, liver triglyceride concentration a, circulating total cholesterol, HDL-C, and LDL-C concentrations b, and serum concentrations of proinflammatory proteins c. n = 6 per group, *p< 0.05, **p < 0.01, ***p < 0.001 compared with control. d Measurement of β-oxidation using C-labeled palmitate in liver. n = 6 per group, *p < 0.05, **p < 0.01 compared with control. e Relative mRNA expression levels of fatty acid oxidative genes in epididymal WAT (left) and the isoproterenol-stimulated lipolytic response in isolated adipocytes (right). n = 6 per group, *p < 0.05, **p < 0.01, ***p < 0.001 compared with control. f Representative images of H&E-stained sections of livers. n = 6 per group. Scale bar, 100 μm. g Measurement of total ketone bodies in fed and fasted (24 h) blood. n = 6 per group, *p < 0.05, **p < 0.01 compared with fasted controls, determined by two-way ANOVA. h Phospho-S6 240/244 Ser levels in liver lysates from ad libitum-fed, fasted (24 h), and re-fed (2 h) HFD-fed mice infected or not (control) with the indicated adenoviruses (left panels). Immunoblot analyses of the mTORC1 signaling pathway using liver lysates (right panels). n = 6 per group regulators of MPK38. To investigate the mechanism biochemical analyses (Figs. 1 and 2). Smads2/3/4 mark- whereby MPK38 activity is regulated by Smads, we edly stimulate ASK1/TGF-β/p53 signaling, whereas examined whether the MPK38-mediated phosphorylation Smad7 inhibits this. However, such effects were not of Smads could affect the activity of MPK38, using various observed in the presence of MPK38-mediated Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 11 of 13 phosphorylation-defective Smad mutants. These findings In conclusion, our data show that Smad proteins differ- indicate that Smad proteins have specific regulatory entially regulate glucose and lipid metabolism and inflam- effects on MPK38 activity, depending on their class, but mation in diet-induced obese mice by differentially also that these effects are strictly dependent on Smad regulating MPK38‐dependent ASK1/TGF-β/p53 signaling, phosphorylation by MPK38. All these results strongly and that this effect is dependent on MPK38-mediated Smad suggest that Smad proteins have novel roles in the dif- phosphorylation and Smad class. Smads2/3/4 function as positive regulators of MPK38, whereas Smad7 functions as ferential regulation of cellular protein kinases. In addition to the effects on MPK38, we also observed a similar trend a negative regulator of MPK38. Moreover, this novel in the effect of Smads on ASK1 activity . function of Smads as differential regulators of MPK38‐ To further enhance understanding of the mechanism(s) dependent ASK1/TGF‐β/p53 signaling provides important by which Smad proteins differentially regulate MPK38 information about the mechanism determining how each activity, we also found that they modify the stability and class of Smad protein contributes to the maintenance of ubiquitination of MPK38. However, such effects were not metabolic homeostasis in mice. detected in MPK38-mediated phosphorylation-defective Smad mutants, indicating that Smad phosphorylation by Materials and methods MPK38 is required for these regulatory effects (Fig. 3). Antibodies, plasmids, chemicals, cell culture, and isolation These results suggest that Smad proteins differentially of hepatocytes and adipocytes regulate complex formation between MPK38 and its known Antibodies and plasmids used for experiments have 7,8,23 regulators, Trx and ZPR9, through direct interactions with been described previously . Cycloheximide (CHX) was MPK38, because Trx and ZPR9 destabilize and stabilize from Sigma-Aldrich. All other chemicals used was 76 252 8,23 MPK38 through Thr and Thr phosphorylation, described . Cell culture and isolation of hepatocytes and 23,24 8,23 respectively . Indeed, the present study demonstrates adipocytes were also described . that Smads2/3/4 decrease MPK38-Trx complex formation and increase MPK38-ZPR9 complex formation, whereas Generation of Smad KI cell lines Smad7 has the opposite effect (Fig. 4). These observations Genomic mutations were generated in HEK293 cells support a model in which Smad proteins differentially using the CRISPR/Cas9 system, as described previously . modify MPK38-Trx and/or MPK38-ZPR9 complex forma- Briefly, single-guide (sg) RNAs were designed to target the tion depending on their class and contribute to the differ- genomic areas adjacent to the Smad mutation sites (Sup- ential regulation of MPK38 kinase activity, resulting in plementary Table 1). Two complementary oligonucleotides specific regulation of MPK38-dependent ASK1/TGF‐β/ (Supplementary Table 2) containing the appropriate Smad p53 signaling pathways (Supplementary Fig. 5). guide sequence and Bbs1 ligation adapters were synthesized Obese mice display lower MPK38 kinase activity , by Bioneer Ltd. (Cheongwon, Korea). The annealed oligo- ASK1/TGF-β/p53 signaling , and Smads2/3/4 expression, nucleotides were ligated into a Bbs1-digested pX458 vector and higher levels of Smad7 expression (Supplementary (Addgene plasmid no. 48138) using the Quick-Ligation Fig. 6), compared with control mice. These results raise system (New England BioLabs). To generate the Smad KI the possibility that Smad proteins are involved in obesity- cell lines, HEK293 cells were cultured on a 24-well plate to associated metabolism, probably by regulating the acti- ~60% confluence and co-transfected with 1 μgSmadsg vation of ASK1/TGF‐β/p53 signaling through MPK38. To RNA plasmid and pUC19 Smad (Smad2 S245A, Smad3 test this hypothesis, we employed an adenoviral delivery S204A, Smad4 S343A, or Smad7 T96A) using Lipofecta- system to restore the lower levels of ASK1/TGF-β/ mine 2000 (Invitrogen). After culturing in a 96-well plate, p53 signaling in obese mice and analyzed the effect of GFP-positive cells were identified, followed by genomic Smads on obesity-associated metabolic abnormalities in DNA extraction. Smads were amplified by PCR using HFD-fed mice. Forced expression of Smads2/3/4, but not Smad-specific PCR primer pairs (Supplementary Table 3). MPK38-mediated phosphorylation-defective Smad The PCR products were A-tailed and cloned into the mutants, induced a significant increase in ASK1/TGF-β/ pGEM-T Easy vector (Promega) to confirm the identity of p53 signaling activation (Supplementary Fig. 4) and individual Smad KI clones by DNA sequencing. The in vivo ameliorated glucose and lipid metabolism (Figs. 5 and 6) phosphorylation of Smads by MPK38 was validated using in HFD-fed mice versus uninfected mice fed a HFD, in vitro kinase assays (Supplementary Fig. 7). whereas Ad-Smad7 infection decreased ASK1/TGF-β/ p53 signaling and further worsened the impaired glucose Coimmunoprecipitation, immunoblotting, MPK38 kinase and lipid metabolism in a MPK38 phosphorylation- assay, apoptosis assay, RNA isolation, and quantitative PCR dependent manner. In addition, MPK38 kinase activity (qPCR) was increased by Smads2/3/4 but decreased by Smad7 in Coimmunoprecipitation, immunoblot analysis, and HFD-fed mice (Supplementary Fig. 4). MPK38 kinase assay were performed as previously Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 12 of 13 described using Smad KI cells or HEK293 cells transiently 10 plaque-forming units) were directly injected into the transfected with the indicated expression vectors or tail vein or epididymal fat pads of 12- to 14-week-old 7,11 Smad-specific siRNAs . Apoptosis and qPCR were HFD-fed mice. 8,11 performed as described previously . Data analysis Glucose and insulin tolerance tests (GTTs/ITTs), Data are expressed as mean ± standard error and are lipogenesis/lipolysis assays, and blood metabolic representative of at least three independent experiments. parameters Statistical significance was determined by one-way or Plasma glucose levels for GTTs/ITTs were measured two-way analysis of variance (ANOVA), followed by using blood obtained from the tail vein at 0, 10, 20, 30, 45, Tukey’s multiple comparison test, using GraphPad Prism 60, 90, and 120 min post-injection according to the software (GraphPad Software). Mouse Metabolic Phenotyping Center recommendations 32 Acknowledgements for data presentation . Lipogenesis and lipolysis were This work was supported by a National Research Foundation of Korea Grant assayed in hepatocytes and adipocytes obtained from (2015R1A2A2A01006098) to H.H. and in part by a National Research Foundation of Korea Grant (2012R1A1A3015539) to H.-A.S. HFD-fed mice infected with the indicated adenoviruses, as described previously . Serum was obtained by cen- Author details trifugation of blood samples collected from the abdominal 1 Department of Biochemistry, School of Biological Sciences, Chungbuk aorta of 12–14-week-old HFD-fed mice and stored at National University, Cheongju 28644, Republic of Korea. Department of Biochemistry, University of Madras, Guindy Campus, Chennai 600025, India −70 °C. Serum levels of tumor necrosis factor (TNF)-α, IL-1β, IL-6, and monocyte chemoattractant protein 1 Authors' contributions (MCP1) were determined using analysis kits from H.-A.S. and R.M. conducted experiments and analyzed data. H.H. designed experiments and wrote the manuscript. Peprotech (ADI-900-047), RayBio (ELM-IL1beta-001), Invitrogen (KMC0061), and R&D Systems (MJE00), Conflict of interest respectively. Other blood metabolic parameters were The authors declare that they have no conflict of interest. measured as described previously . Publisher's note Serum insulin and glucose, liver triglyceride, 2-deoxy- Springer Nature remains neutral with regard to jurisdictional claims in glucose uptake, total ketone, and fatty acid β-oxidation published maps and institutional affiliations. Serum insulin and glucose levels, liver triglyceride Supplementary Information accompanies this paper at https://doi.org/ content, and 2-deoxy-glucose uptake in epididymal WAT 10.1038/s41419-018-0489-x. and soleus muscles were measured as described pre- 8,33 viously . Total ketone levels were measured using a Received: 3 October 2017 Revised: 15 March 2018 Accepted: 19 March colorimetric assay from Wako Chemicals . Fatty acid oxidation rate was determined using [1- C] palmitic acid, as described previously . Animal experiments and adenoviral infection References Four-week-old male C57BL/6 mice purchased from 1. Wrighton, K. H., Lin, X. & Feng, X. H. Phospho-control of TGF-beta superfamily Orient (Seongnam, Korea) were fed a HFD (60% kcal as signaling. Cell Res. 19,8–20 (2009). 2. Bierie, B. & Moses, H. L. Tumour microenvironment: TGFbeta: the molecular fat diet, D12492; Research Diets, Inc.) for 8–10 weeks Jekyll and Hyde of cancer. Nat. Rev. Cancer 6,506–520 (2006). under a 12:12-h light:dark cycle. Male 12- to 14-week-old 3. Massagué, J., Blain, S. W. & Lo, R. S. TGFbeta signaling in growth control, cancer, and heritable disorders. Cell 103,295–309 (2000). C57BL/6 mice fed with a standard rodent diet were 4. Siegel,P.M., Shu, W.,Cardiff, R.D., Muller, W. J. &Massagué, J. Transforming used as controls. All procedures were performed in a growth factor beta signaling impairs Neu-induced mammary tumorigenesis specific pathogen-free facility at Chungbuk National while promoting pulmonary metastasis. Proc. Natl. Acad. Sci. USA 100, 8430–8435 (2003). University, in accordance with the approved animal pro- 5. Massagué,J.TGFbeta in. Cancer Cell. 134,215–230 (2008). tocols and the guidelines established by the Ethics Review 6. Li, M. O. & Flavell, R. A. TGF-beta: a master of all T cell trades. Cell 134,392–404 Committee of Chungbuk National University for Animal (2008). 7. Seong, H. A., Jung, H. & Ha, H. Murine protein serine/threonine kinase 38 sti- Experiments (CBNUA-966-16-02). Recombinant adeno- mulates TGF-beta signaling in a kinase-dependent manner via direct phos- viruses were generated as previously described .To phorylation of Smad proteins. J. Biol. Chem. 285, 30959–30970 (2010). prepare adenoviruses expressing wild-type and MPK38 8. Seong, H. A.,Manoharan,R.& Ha, H. Coordinateactivation ofredox- dependent ASK1/TGF-β signaling by a multiprotein complex (MPK38, ASK1, phosphorylation-defective Smad mutants, FLAG-tagged SMADs, ZPR9, and TRX) improves glucose and lipid metabolism in mice. wild-type and mutant Smad plasmids were used as tem- Antioxid. Redox Signal. 24,434–452 (2016). plates for PCR with the primers shown in Supplementary 9. Saito, R., Tabata, Y., Muto, A., Arai, K. & Watanabe, S. Melk-like kinase plays a role Table 4. Recombinant adenoviruses (approximately 1 × in hematopoiesis in the zebra fish. Mol. Cell. Biol. 25,6682–6693 (2005). Official journal of the Cell Death Differentiation Association Seong et al. Cell Death and Disease (2018) 9:471 Page 13 of 13 10. Beullens, M. et al. Substrate specificity and activity regulation of protein kinase 23. Seong, H. A., Manoharan, R. & Ha, H. Zinc finger protein ZPR9 functions as an MELK. J. Biol. Chem. 280, 40003–40011 (2005). activator of AMPK-related serine/threonine kinase MPK38/MELK involved in 11. Jung, H., Seong, H. A. & Ha, H. Murine protein serine/threonine kinase 38 ASK1/TGF-β/p53 signaling pathways. Sci. Rep. 7, 42502 (2017). activates apoptosis signal-regulating kinase 1 via Thr 838 phosphorylation. 24. Manoharan, R., Seong, H. A. & Ha, H. Thioredoxin inhibits MPK38-induced J. Biol. Chem. 283, 34541–34553 (2008). ASK1, TGF‐β, and p53 function in a phosphorylation-dependent manner. Free. 12. Gray, D. et al. Maternal embryonic leucine zipper kinase/murine protein serine- Radic. Biol. Med. 63,313–324 (2013). threonine kinase 38 is a promising therapeutic target for multiple cancers. 25. Tan,C.K., Chong, H. C.,Tan,E.H.&Tan, N. S. Getting ‘Smad’ about obesity and Cancer Res. 65, 9751–9761 (2005). diabetes. Nutr. Diabetes 2, e29 (2012). 13. Vulsteke, V. et al. Inhibition of spliceosome assembly by the cell cycle- 26. Berkers,C.R., Maddocks,O. D., Cheung,E. C., Mor, I. & Vousden,K. H. Metabolic regulated protein kinase MELK and involvement of splicing factor NIPP1. J. Biol. regulation by p53 family members. Cell. Metab. 18,617–633 (2013). Chem. 279,8642–8647 (2004). 27. van Diepen, J. A., Berbée, J. F., Havekes, L. M. & Rensen, P. C. Interactions 14. Higaki, M. & Shimokado, K. Phosphatidylinositol 3-kinase is required for growth between inflammation and lipid metabolism: relevance for efficacy of anti- factor-induced amino acid uptake by vascular smooth muscle cells. Arter. inflammatory drugs in the treatment of atherosclerosis. Atherosclerosis 228, Thromb. Vasc. Biol. 19,2127–2132 (1999). 306–315 (2013). 15. Seong, H. A., Jung, H., Kim, K. T. & Ha, H. 3-Phosphoinositide-dependent PDK1 28. Muoio,D.M. & Newgard, C. B. Molecular and metabolic mechanisms of insulin negatively regulates transforming growth factor-beta-induced signaling in a resistance and β-cell failure in type 2 diabetes. Nat. Rev. Mol. Cell. Biol. 9, kinase-dependent manner through physical interaction with Smad proteins. 193–205 (2008). J. Biol. Chem. 282, 12272–12289 (2007). 29. Sengupta, S., Peterson, T. R., Laplante, M., Oh, S. & Sabatini, D. M. mTORC1 16. Tan,C.K. etal. Smad3deficiency in mice protects against insulin resistance controls fasting-induced ketogenesis and its modulation by ageing. Nature and obesity induced by a high-fat diet. Diabetes 60,464–476 (2011). 468, 1100–1104 (2010). 17. Yadav, H. et al. Protection from obesity and diabetes by blockade of TGF-β/ 30. Blüher, M. et al. Activated Ask1-MKK4-p38MAPK/JNK stress signaling pathway Smad3 signaling. Cell. Metab. 14,67–79 (2011). in human omental fat tissue may link macrophage infiltration to whole-body 18. Tsurutani, Y. et al. The roles of transforming growth factor-β and Insulin sensitivity. J. Clin. Endocrinol. Metab. 94, 2507–2515 (2009). Smad3 signaling in adipocyte differentiation and obesity. Biochem. Biophys. 31. Hattori, K. et al. ASK1 signalling regulates brown and beige adipocyte function. Res. Commun. 407,68–73 (2011). Nat. Commun. 7, 11158 (2016). 19. Lin, H. M. et al. Transforming growth factor-beta/Smad3 signaling regulates 32. McGuinness, O. P.,Ayala,J.E., Laughlin,M.R.&Wasserman, D. H. NIH insulin gene transcription and pancreatic islet beta-cell function. J. Biol. Chem. experiment in centralized mouse phenotyping: the Vanderbilt experience 284, 12246–12257 (2009). and recommendations for evaluating glucose homeostasis in the mouse. 20. Perry, J. R. et al. Interrogating type 2 diabetes genome-wide association data Am.J.Physiol.Endocrinol. Metab. 297, E849–E855 (2009). using a biological pathway-based approach. Diabetes 58,1463–1467 (2009). 33. Chang, P. Y. et al. Overexpression of hexokinase II in transgenic mice: evidence 21. Nomura, M. et al. SMAD2 disruption in mouse pancreatic beta cells leads to that increased phosphorylation augments muscle glucose uptake. J. Biol. islet hyperplasia and impaired insulin secretion due to the attenuation of ATP- Chem. 271,14834–14839 (1996). sensitive K+ channel activity. Diabetologia 57, 157–166 (2014). 34. Seong, H. A., Manoharan, R. & Ha, H. A crucial role for the phosphorylation of 22. Li, H. Y. et al. Amelioration of high fat diet-induced glucose intolerance by STRAP at Ser(188) by MPK38 in STRAP-dependent cell death through ASK1, blockade of Smad4 in pancreatic beta-cells. Exp. Clin. Endocrinol. Diabetes 123, TGF-β, p53, and PI3K/PDK1 signaling pathways. Cell Cycle 13,3357–3374 221–226 (2015). (2014). Official journal of the Cell Death Differentiation Association

Journal

Cell Death & DiseaseSpringer Journals

Published: Apr 27, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off