Atrophic Remodeling of the Heart In Vivo Simultaneously Activates Pathways of Protein Synthesis and Degradation
Background— Mechanical unloading of the heart results in atrophic remodeling. In skeletal muscle, atrophy is associated with inactivation of the mammalian target of rapamycin (mTOR) pathway and upregulation of critical components of the ubiquitin proteosome proteolytic (UPP) pathway. The hypothesis is that mechanical unloading of the mammalian heart has differential effects on pathways of protein synthesis and degradation.
Methods and Results— In a model of atrophic remodeling induced by heterotopic transplantation of the rat heart, we measured gene transcription, protein expression, polyubiquitin content, and regulators of the mTOR pathway at 2, 4, 7, and 28 days. In atrophic hearts, there was an increase in polyubiquitin content that peaked at 7 days and decreased by 28 days. Furthermore, gene and protein expression of UbcH2, a ubiquitin conjugating enzyme, was also increased early in the course of unloading. Transcript levels of TNF-α, a known regulator of UbcH2-dependent ubiquitin conjugating activity, were upregulated early and transiently in the atrophying rat heart. Unexpectedly, p70S6K and 4EBP1, downstream components of mTOR, were activated in atrophic rat heart. This activation was independent of Akt, a known upstream regulator of mTOR. Rapamycin treatment of the unloaded rat hearts inhibited the activation of p70S6K and 4EBP1 and subsequently augmented atrophy in these hearts compared with vehicle-treated, unloaded hearts.
Conclusions— Atrophy of the heart, secondary to mechanical unloading, is associated with early activation of the UPP. The simultaneous activation of the mTOR pathway suggests active remodeling, involving both protein synthesis and degradation.
Received March 7, 2003; de novo received June 12, 2003; revision received July 18, 2003; accepted July 22, 2003.
Atrophy is characterized by a change in the balance of protein turnover in favor of proteolysis. The ubiquitin-dependent pathway of protein degradation is responsible for the major part of protein breakdown in skeletal muscle.1 Previous studies in skeletal muscle atrophy have shown that activation of the UPP is associated with an increase in mRNA levels of ubiquitin, ubiquitin conjugating enzymes, ubiquitin ligases, and components of the proteasome.2–5 The mTOR pathway is thought to be the main signaling cascade that activates protein translation, thereby regulating protein synthesis.6 Atrophy in skeletal muscle leads to the decreased phosphorylation of downstream proteins of mTOR, such as p70S6K.7 Thus, in skeletal muscle, atrophy protein degradation is activated whereas protein synthesis is inhibited.
Mechanical unloading of the heart leads to atrophic remodeling with reactivation of the fetal gene program.8 We wanted to know how similar the processes of atrophic remodeling in heart and skeletal muscle were. Thus, we used a model of mechanical unloading with heterotopic rat heart transplantation to investigate atrophy in the heart. We studied the time course by analyzing heart tissue at 2, 4, 7, and 28 days after the heterotopic transplantation of the rat heart.
We found increased abundance of polyubiquitin protein content in unloaded rat hearts. In addition, gene and protein expression of UbcH2, a ubiquitin-conjugating enzyme, was also increased early in the course of unloading. Transcript levels of TNF-α, a known regulator of UbcH2-dependent ubiquitin conjugating activity,9 were also upregulated early and transiently in the atrophying rat heart. In contrast to skeletal muscle,7 we observed increased phosphorylation of p70S6K and 4EBP1, which was rapamycin-sensitive but independent of Akt/PKB, a known upstream regulator of mTOR. Unloaded rat hearts treated with rapamycin atrophied to a greater extent than vehicle-treated hearts. Collectively, our findings suggest that the UPP is rapidly activated in the unloaded heart. Furthermore, mTOR is also activated in atrophic remodeling of the heart and probably serves as a feedback mechanism to maintain heart size.
Male Wistar rats (160 to 200 g) underwent heterotopic heart transplantation as previously described.8 Briefly, isogenic infrarenal heterotopic heart transplantation was performed by anastomizing end-to-side the ascending aorta of the donor to the abdominal aorta of the recipient and the donor pulmonary artery to the recipient inferior vena cava. In some transplanted rats, rapamycin at 0.4 mg/kg per day or vehicle was administered for 7 days by osmotic pump before sacrifice as previously described.10 Briefly, rapamycin was obtained at concentration of 19.66 mg/mL and diluted to 0.4 mg/kg per day, with the use of a dilution of Tween 80 (10%), N-N-dimethylacetamide (20%), and polyethylene glycol. A dose-response curve for rapamycin was done in heterotopically transplanted rat hearts that demonstrated that 0.32 mg/kg per day is an effective dose. The use of animals and the animal protocol were approved by the Animal Welfare Committee of the University of Texas Houston Health Science Center.
At 2, 4, 7, and 28 days after surgery, animals were anesthetized (pentobarbital, 100 mg/kg body wt IP) and donor and recipient hearts were rapidly removed, freeze-clamped, and stored at −80°C. Tissue was immediately frozen in liquid nitrogen for RNA and protein extraction.
RNA was extracted by standard methods and analyzed by reverse transcription followed by real-time quantitative PCR for the transcripts of interest by methods described previously.8 Nucleotide sequences for probes as well as forward and reverse primers of the human transcript are shown in the Table. The transcript for the housekeeping gene 18S was used as a reporter gene for data normalization of rat samples. Gene expression of 18S did not significantly differ between the groups (data not shown). Internal RNA standards were prepared with the use of the T7 RNA polymerase method (Ambion).
Protein from the native and heterotopic rat hearts was isolated as previously described.8 Briefly, samples were homogenized in extraction buffer and centrifuged (30 minutes at 15 000g), and the supernatant was isolated. Immunoprecipitation was performed by adding antibody for ubiquitin (Santa Cruz sc-9133) to supernatant at concentration of 1:1000 and rotating in a cold room overnight. Protein A Sepharose beads were added to the supernatant and allowed to rock at 4°C for 1 hour. The mixture was centrifuged. The pellet was then washed and centrifuged 3 times with PBS. Fifty microliters of 2× SDS loading buffer was added to each sample and boiled for 5 minutes to denature protein. Samples were fractionated by 6% PAGE and transferred to a PVDF membrane. Antibody for polyubiquitin (Santa Cruz sc-8017) was added at concentration of 1:100 in 3% milk and shaken in the cold room overnight. Anti-mouse IgG-HRP (Santa Cruz sc-2005) was then added at concentration of 1:2000 in 3% milk and shaken for 2 hours at room temperature. Five milliliters of Luminal A and B reagents (Santa Cruz sc-2048) was added to membrane and exposed to film for 2 minutes. Western blots were performed with the antibodies for p70S6K, 4EBP1, and Akt1/2 (Santa Cruz sc-8418, sc-6024, and sc-8312, respectively). The antibodies for phospho-Akt1/2 (at ser 473) and phospho-p70S6K (at Thr 389) were purchased from Cell Signaling (9271S and 92055, respectively). We used antibody for GAPDH to normalize for protein loading (Research Diagnostics). The antibody for UbcH2 is derived from rabbit antiserum raised against recombinant UbcH2 as described elsewhere.9
Data are expressed as mean±SEM. Differences between the groups were calculated by a 1-way ANOVA followed by a Bonferroni test. A value of P<0.05 was considered significant.
The weight of unloaded, heterotopically transplanted rat hearts as a percentage of native heart weight over time is shown in Figure 1A (n=5 for each time point). The rate of atrophy of the unloaded heart was highest between 2 and 4 days and decreased soon after.
To explore whether the decrease in heart weight of the unloaded rat hearts was associated with activation of the UPP, we performed immunoprecipitation and protein gel electrophoresis studies by using antibodies for polyubiquitin. Figure 1B shows a marked increase in polyubiquitin content in a representative protein gel (n=6). Densitometry was performed on the 6 gels and graphed in Figure 1C. Polyubiquitin content increased as early as 2 days (P<0.05) after transplantation, peaked around 7 days (P<0.01), and trended downward by 28 days (P<0.05).
Gene Expression of Ubiquitin-Dependent Enzymes
We analyzed several key ubiquitin-conjugating enzymes that are known to be involved in skeletal muscle atrophy. There is a significant increase (P<0.05) in the expression of the ubiquitin-conjugating enzyme UbcH2 seen at 2 days after transplantation that rapidly returns to baseline by 4 days (Figure 2A). Transcript levels of other key regulatory enzymes (E2–14k, MAFbx/Atrogin1) did not change significantly (data not shown).
Protein Expression of UbcH2
Changes in UbcH2 mRNA transcript levels in the unloaded rat hearts were paralleled by similar changes in UbcH2 protein levels. Figure 2B shows an increase in UbcH2 in the day-2 transplanted hearts; protein levels remained elevated up to 7 days and returned to baseline by 28 days (n=3).
Gene Expression of TNF-α
To determine if UbcH2 expression was regulated by TNF-α in the heart, we measured the gene expression of TNF-α in the unloaded rat hearts. Transcript levels of TNF-α were increased in the atrophying rat hearts at 2 days and returned to baseline by 4 days (P<0.05, Figure 2C).
mTOR Signaling Pathway
Unexpectedly, p70S6K and 4EBP1 were phosphorylated in response to mechanical unloading (Figure 3A). This activation was seen early and was sustained up to 28 days and was independent of Akt/PKB phosphorylation (Figure 3B).
Rapamycin Augments Atrophy in the Unloaded Hearts
Rapamycin at 0.4 mg/kg per day or vehicle was administered by osmotic pump to heterotopically transplanted rat hearts for 7 days. Rapamycin prevented the phosphorylation of p70S6K and 4EBP1 observed with unloading (Figure 3A). Treatment with rapamycin resulted in a nearly 15% decrease in heart weight of the unloaded hearts (P<0.05, Figure 3C).
The main finding of this study is the simultaneous activation of pathways regulating proteolysis and protein synthesis in atrophic remodeling of the rat heart. We demonstrate upregulation of markers of the UPP as well as activation of downstream components of mTOR. TNF-α expression, a regulator of UbcH2-dependent conjugation activity, was also increased early in the course of unloading. Rapamycin not only prevented the activation of mTOR in unloading but also augmented atrophy in transplanted hearts by nearly 15%.
Atrophy of the Heart Parallels Upregulation of the UPP
Mechanical unloading of rat hearts with heterotopic transplantation resulted in an atrophic remodeling of the myocardium that was most rapid in the first 2 days of unloading, peaked between 2 and 4 days, and slowed by 28 days. This change in the rate of the trophic response was paralleled by changes of polyubiquitin content, gene expression, and protein content of the ubiquitin-conjugating enzyme UbcH2.
Transient changes in markers of the UPP in cardiac atrophy are consistent with studies examining skeletal muscle atrophy. In a rat model of sepsis-induced atrophy of skeletal muscle, transcript levels of ubiquitin and a proteosome subunit increased after 2 days but normalized after 10 days of sepsis.3 In contrast, gene expression of two key regulatory enzymes (E2–14k, MAFbx1/Atrogin1) that increase during skeletal muscle atrophy11 did not significantly increase in the atrophied rat heart.
Although the most rapid decline in heart weight occurred between 2 and 4 days in our study, polyubiquitin content peaked at 7 days. We speculate that rapid ubiquitin conjugation may have resulted in ubiquitinated proteins accumulating faster than they could be degraded by the proteosome. As the rate of atrophy slows down, a new steady state is reached and there is tighter coupling between ubiquitin conjugation and proteosomal degradation. Thus, the increased polyubiquitin content at 7 days may reflect a “time lag” caused by a discrepancy between ubiquitin conjugation and proteosomal degradation.
UbcH2 Expression Parallels TNF-α Expression
Recently, UbcH2 was found to play a key role in TNF-α stimulated protein degradation in skeletal muscle.9 TNF-α, formerly known as cachexin, is thought to stimulate skeletal muscle atrophy in a number of diseases, including heart failure.12–15 In our study, expression of UbcH2 in the atrophying heart was paralleled by expression of TNF-α. TNF-α expression in the heart can be transiently induced by increasing hemodynamic load. For example, transverse aortic constriction in mice resulted in an early increase in TNF-α expression, which returned to baseline levels by 72 hours.16 We discovered a similar transient upregulation of TNF-α transcript levels with mechanical unloading. We postulate that mechanical unloading of the heart may alter wall stress and/or modify the stretch properties of cardiac myocytes resulting in increased expression of TNF-α. TNF-α then activates the UPP (characterized by upregulation of UbcH2), resulting in atrophy. We cannot, however, rule out the possibility that transient upregulation of TNF-α may result from an inflammatory reaction of the transplanted heart. Our findings also suggest that the UPP may have a role in hypertrophic remodeling of the mammalian heart. We hope to characterize the molecular mechanisms underlying TNF-α–stimulated atrophy and remodeling in future investigations.
Activation of mTOR in Atrophy
The phosphorylation of p70S6K and 4EBP1 in atrophy of the rat heart is an unexpected finding. The activation of both proteins was rapamycin-sensitive, indicating that upstream signaling is through the mTOR pathway. This pathway, which regulates protein synthesis, is activated in both cardiac and skeletal muscle hypertrophy.7,17,18 Furthermore, there is evidence of inactivation of components of the mTOR signaling cascade in skeletal muscle atrophy.7 Akt/PKB is also thought to be an upstream regulator of mTOR.7,17 However, in our study, there was no change in the phosphorylation of Akt/PKB, and, therefore, activation of mTOR probably is independent of the insulin/IGF-1 signaling cascade. Intracellular amino acids can activate mTOR through an unknown mechanism.19 We have previously shown that glutamine administration causes hypertrophy of cardiomyocytes with activation of mTOR.20 We speculate that atrophic remodeling of the heart may result in an accumulation of intracellular amino acids secondary to breakdown by the UPP, which subsequently leads to the activation of mTOR.
Activation of mTOR May Regulate Proteolysis
We speculate that the activation of signaling pathways that regulate protein synthesis in mechanical unloading may represent a feedback mechanism to prevent excessive atrophy of the heart. In cardiac atrophy induced by protein caloric malnutrition, the heart atrophies less than other organs and cardiac mass is restored earlier with nutritional replenishment, suggesting that regulation of protein metabolism differs in the heart.21 Amino acids are thought to regulate proteolysis by autophagy in multiple organs through activation of mTOR-mediated protein translation.22 The observation in this study that rapamycin treatment for 7 days augments atrophy by nearly 15% supports this speculation.
Our data appear to be in conflict with those of Klein et al,23 who found that the capacity for protein synthesis is decreased in heterotopically transplanted rat hearts. However, the same authors found that the efficiency of protein synthesis was unchanged,23 a finding consistent with our observations in this study. It is likely that mTOR activation is required for basal protein synthesis during atrophy and that inhibition of this pathway by rapamycin blocks normal steady-state protein synthesis, thereby accelerating atrophy.
The interaction between protein synthesis and degradation may also play an important role in the hypertrophic remodeling of the heart. Protein synthesis is obviously increased in the hypertrophying heart. It has also been shown that there is an increased turnover rate of existing proteins when the heart is subjected to increased work load.24 Thus, the concept of “dynamic nature of body constituents”25 applies to both atrophic and hypertrophic remodeling of the heart. The latter is a subject of ongoing investigation in our laboratory.
Atrophic remodeling of the heart is associated with simultaneous activation of regulators of both proteolysis and protein synthesis. The UPP is upregulated early and transiently with unloading, whereas mTOR activation is more sustained. The activation of mTOR may represent a feedback mechanism that prevents excessive atrophy and maintains heart size.
This study was supported in part by grants from the National Heart, Lung, and Blood Institute (RO1-HL/AG 61483 and T32-HL 07591 to H.T. and RO1-HL 59878 to M.B.R.) and the American Heart Association, National Center.
Guest Editor for this article was Joseph Loscalzo, MD, PhD, Boston University School of Medicine, Boston, Mass.
↵*These authors contributed equally to this work.
Lecker S, Soloman V, Mitch W, et al. Muscle protein breakdown and the critical role of the ubiquitin-proteosome pathway in normal and disease states. J Nutr. 1999; 129: 227S–237S.
Bodine SC, Latres E, Baumhueter S, et al. Identification of ubiquitin ligases required for skeletal muscle atrophy. Science. 2001; 294: 1704–1708.
Medina R, Wing SS, Goldberg AL. Increase in levels of polyubiquitin and proteasome mRNA in skeletal muscle during starvation and denervation atrophy. Biochem J. 1995; 307: 631–637.
Li Y, Lecker S, Chen Y, et al. TNF-alpha increases ubiquitin conjugating activity in skeletal muscle by upregulating UbcH2/E220K. FASEB J. 2003; 17: 1048–1057.
Gomes M, Lecker S, Jagoe R, et al. Atrogin-1, a muscle-specific F-box protein highly expressed during muscle atrophy. Proc Natl Acad Sci U S A. 2001; 98: 14440–14445.
Tisdale M. Wasting in cancer. J Nutr. 1999; 129: 243S–246S.
Baumgarten G, Knuefermann P, Kalra D, et al. Load-dependent and -independent regulation of proinflammatory cytokine and cytokine receptor gene expression in the adult mammalian heart. Circulation. 2002; 105: 2192–2197.
Shioi T, McMullen J, Tarnavski O, et al. Rapamycin attenuates load-induced cardiac hypertrophy in mice. Circulation. 2003; 107: 1664–1670.
Pallafacchina G, Calabria E, Serrano E, et al. A protein kinase B-dependent and rapamycin-sensitive pathway controls skeletal muscle growth but not fiber type specification. Proc Natl Acad Sci U S A. 2002; 99: 9213–9218.
Xia Y, Wen HY, Young ME, et al. mTOR and PKA signaling mediate the cardiac transcriptional response to glutamine. J Biol Chem. 2003; 278: 13143–13150.
Kadowaki M, Kanazawa T. Amino acids as regulators of proteolysis. J Nutr. 2003; 133: 2052S–2056S.
Klein I, Samarel AM, Welikson R, et al. Heterotopic cardiac transplantation decreases the capacity of rat myocardial protein synthesis. Circ Res. 1991; 68: 1100–1107.
Morkin E, Kimata S, Skillman JJ. Myosin synthesis and degradation during development of cardiac hypertrophy in the rabbit. Circ Res. 1972; 30: 690–702.
Schoenheimer R. The Dynamic State of Body Constituents. Cambridge, Mass: Harvard University Press; 1942.