Differential control of muscle mass in type 1 and type 2 diabetes mellitus

Abstract

Diabetes mellitus—whether driven by insulin deficiency or insulin resistance—causes major alterations in muscle metabolism. These alterations have an impact on nutrient handling, including the metabolism of glucose, lipids, and amino acids, and also on muscle mass and strength. However, the ways in which the distinct forms of diabetes affect muscle mass differ greatly. The most common forms of diabetes mellitus are type 1 and type 2. Thus, whereas type 1 diabetic subjects without insulin treatment display a dramatic loss of muscle, most type 2 diabetic subjects show no changes or even an increase in muscle mass. However, the most commonly used rodent models of type 2 diabetes are characterized by muscle atrophy and do not mimic the features of the disease in humans in terms of muscle mass. In this review, we analyze the processes that are differentially regulated under these forms of diabetes and propose regulatory mechanisms to explain them.

Introduction

Diabetes mellitus is a worldwide health problem that currently affects 387 million people, representing a prevalence of 8.3 % of the total population [1]. The term diabetes mellitus refers to a group of pathologies that are characterized and diagnosed by high levels of circulating glucose caused by compromised insulin function, either because this hormone is absent or because its action is inefficient [2, 3]. Diabetes mellitus is classified into types on the basis of the underlying cause of altered insulin function, type 1 and type 2 being the most common manifestations [2, 3].

Of all the individuals with diabetes mellitus, 5–10 % have type 1, the incidence of which is increasing yearly [4, 5]. Type 1 diabetes is an autoimmune disease characterized by the absence of insulin in the organism. β-Cells, which are the pancreatic cells in charge of the production and release of insulin, are destroyed by the host immune system [4]. Both genetic and environmental factors have been related to the development of type 1 diabetes [6–8], and patients need insulin treatment in order to survive [2].

Type 2 diabetes mellitus is the most common form of this disease, accounting for approximately 90 % of cases diagnosed [2]. This type of diabetes is characterized by insulin resistance—which means that peripheral tissues do not respond properly to the presence of this hormone—together with impairment of insulin secretion as the disease progresses [2, 9–12]. The onset of type 2 diabetes is a long process, initially characterized by the development of insulin resistance, which is at first compensated by increased production of this hormone by β-cells. However, over time, insulin secretion is inadequate and unable to compensate for insulin resistance [2].

The development of insulin resistance and type 2 diabetes has been associated with several factors. Type 2 diabetes mellitus is considered a complex disease with a genetic component [13–15]. More than 50 genetic loci associated with type 2 diabetes have been identified, thus explaining a modest fraction of heritability [16]. Environmental factors are also crucial for the development of this pathology, and type 2 diabetes and insulin resistance have been directly related to obesity and a sedentary lifestyle [10, 17–19].

Insulin resistance affects several peripheral tissues, the liver, adipose tissue, and skeletal muscle being the most relevant for the development of the associated alterations in whole-body metabolism. Traditionally, the alterations in glucose and lipid metabolism in type 2 diabetes have received the most attention. However, the impact of diabetes mellitus on protein metabolism and skeletal muscle mass has gained interest in recent years. In fact, skeletal muscle, which is not only the major tissue in charge of glucose uptake upon insulin stimulation but also the main protein reservoir of the organism, is one of the tissues most severely affected by this disease [11].

With respect to the impact of diabetes mellitus on skeletal muscle mass, type 1 and type 2 diabetes show marked differences. Both pathologies display compromised insulin function, but due to distinct underlying causes. While type 1 diabetic subjects without treatment display dramatic muscle loss [20–25], most type 2 diabetic subjects show no changes or an increase in muscle mass (except in elderly subjects in which type 2 diabetes favors sarcopenia) [26–31].

On the basis of these considerations, in this review we will analyze the effects of type 1 and type 2 diabetes mellitus on skeletal muscle mass and protein metabolism and the potential mechanisms that account for the differences between these two pathological conditions.

Mechanisms regulating skeletal muscle mass

Skeletal muscle has a remarkable capacity to adapt and respond to a huge variety of stimuli, such as physical activity, nutrient availability, circulating levels of hormones, cytokines, and growth factors [32]. Such stimuli can affect muscle mass and myofiber size by changing the balance between protein synthesis and protein degradation [32, 33]. Various cellular systems, including the Ubiquitin Proteasome System (UPS), the lysosomal system, calpains, and caspases, are involved in protein degradation. In most cell types, including skeletal muscle, the UPS degrades most of the intracellular proteins [34]. Briefly, protein degradation through the UPS is a sequential process in which the enzymes E1, E2, and E3 conjugate ubiquitin to the target to be degraded. After the addition of this first ubiquitin to the substrate, more ubiquitin molecules are added, thus leading to the formation of a polyubiquitin chain. Once a chain of four ubiquitin molecules or more has formed, this protein is recognized and degraded by the 26S proteasome [35]. Protein degradation through this process requires energy provided by ATP hydrolysis [36]. Based on experiments performed in rats, protein degradation via the UPS represents around 50 % of total protein degradation in skeletal muscle, although differences between muscle types have been observed. For example, a lower percentage of UPS-dependent protein degradation has been reported in soleus compared to EDL (Extensor Digitorum Longus) or diaphragm [37–39].

Macroautophagy (hereafter referred to as autophagy) has recently been reported to be a crucial player in the regulation of muscle mass and function. Autophagy involves the formation of double-membrane vesicles, known as autophagosomes, which sequester a part of the cytosol (which includes proteins, organelles, etc.) [40]. Autophagosomes fuse with lysosomes to form autolysosomes, and all the cargo inside the autophagosomes is degraded by lysosomal hydrolases [40]. Analysis of the global rates of muscle protein degradation in the absence or presence of lysosomal inhibitors in rats suggests that autophagy may represent between 25 and 30 % of total protein degradation in skeletal muscle, although these data could vary among different muscles [38, 39, 41]. For example, lysosomal protein degradation can represent from the 20 % in epitrochlearis muscle to the 25–35 % in the whole hindquarter [38, 39, 41].

Activation of the UPS and autophagy has been associated with muscle loss in several pathologies, such as sepsis, chronic kidney disease, hepatic cirrhosis, cancer cachexia, and insulinopenic diabetes [21, 22, 42–47]. It has been reported that the UPS and autophagy can be coordinately regulated to promote muscle atrophy [48, 49]. Moreover, several conditions of muscle wasting are characterized by common transcriptional changes in skeletal muscle [50]. Some genes are upregulated while others are downregulated, but always in the same direction in the pathological conditions tested [50]. This group of genes has been called atrogenes, and among the most upregulated ones are those involved in protein degradation through the UPS (i.e., MuRF1, atrogin 1) and autophagy (i.e., LC3, Bnip3) [48–52]. Insulin has been described to inhibit protein degradation in human skeletal muscle [23, 25, 53–55]. On the basis of studies in murine models, the effect of insulin on protein catabolism has been classically attributed to the inhibition of the UPS caused by repression of the expression of genes involved in this process [46, 50, 56]. It has also recently been reported that the lack of insulin enhances muscle autophagy in streptozotocin-induced diabetes [57]. Moreover, insulin is able to stimulate protein synthesis in skeletal muscle of murine models and muscle cells in vitro [58–61]. However, this hormone does not appear to affect protein synthesis in human skeletal muscle [23, 25, 53–55]. Overall, current data suggest that insulin influences protein metabolism in human skeletal muscle through inhibiting protein degradation.

Insulin signaling in skeletal muscle

The action of insulin is dependent on its binding to its receptor in the sarcolemma and it requires the activation of its tyrosine kinase activity (Fig. 1). Autophosphorylation of the receptor generates binding sites for IRSs (Insulin Receptor Substrates). While IRS-1 ablation compromises growth in mice [62, 63], IRS-2 ablation does not cause any major defect in growth but these mice develop type 2 diabetes [64]. These data suggest that IRS-1 is a key factor in the determination of cell size. Focusing on skeletal muscle, both IRS-1 and IRS-2 have been described to be relevant for the maintenance of muscle mass [65].

Fig. 1.

Insulin signaling and protein metabolism in skeletal muscle. Schematic view of the effectors that participate in the insulin signaling pathway. Insulin binding to its receptor activates its tyrosine kinase activity and promotes IRS recruitment. Then IRS is phosphorylated and this event generates binding sites for the class I PI3K. PI3K activity generates phosphoinositides that act as binding sites for the recruitment of AKT to the membrane. There, AKT is activated by phosphorylation of other kinases, such as PDK1, and participates in various signaling pathways to promote protein synthesis and inhibit protein degradation. AKT phosphorylates and inhibits GSK3β, a negative regulator of protein synthesis. In addition, AKT phosphorylation also inhibits TSC2, a protein found in a complex with TSC1. The TSC1-TSC2 complex inhibits Rheb by promoting its GTPase activity. Inhibition of TSC2 preserves Rheb function and activates mTORC1. mTORC1 promotes protein synthesis by phosphorylating and inhibiting the 4EBPs, as well as by activating the S6 kinases through phosphorylation. Moreover, mTORC1 inhibits protein degradation by inhibiting autophagy. Finally, AKT also phosphorylates and inhibits the FoxO transcription factors. The FoxOs promote protein degradation by increasing the expression of genes involved in autophagy and in the UPS

Once the corresponding IRS has been recruited, it is phosphorylated by the receptor, thus generating binding sites for class I phosphoinositide 3-kinase (PI3K). The 3-phosphorylated phosphoinositides produced act as a binding site for the recruitment of phosphoinositide-dependent kinase-1 (PDK1) and protein kinase B (PKB or AKT) to the membrane [66]. Once in the membrane, AKT is phosphorylated by at least two kinases, namely PDK1 and mammalian target of rapamycin complex 2 (mTORC2) [67–69]. The phosphorylation of these two residues is necessary for the maximal activity of AKT [67–69], a serine/threonine kinase that acts in several substrates inside the cell. In fact, three isoforms of AKT have been described: AKT1, AKT2, and AKT3. Of these, the most abundant ones in skeletal muscle are AKT1 and AKT2 [70, 71]. AKT1 ablation impairs growth in mice without altering glucose metabolism [72, 73]. In contrast, mice with AKT2 ablation do not show defects in growth but display insulin resistance [73]. The double knock-out model presents the most severe phenotype, with severe impairment of growth, reduced life span, and skeletal muscle atrophy [74]. Overall, these data suggest that AKT2 has a key role in glucose metabolism, while AKT1 is a regulator of cell growth. Thus, the expression of a constitutively active form of AKT1 specifically in skeletal muscle causes muscle hypertrophy [75].

Mammalian target of rapamycin (mTOR) is a serine/threonine kinase found in two complexes in mammalian cells, namely mammalian target of rapamycin complex 1 (mTORC1) and mTORC2. mTORC2 is a complex that phosphorylates AKT, and this phosphorylation is crucial for achieving the maximal activity of AKT, as well as for increasing its stability [67, 68]. However, the role of mTORC2 in AKT activation in skeletal muscle is rather controversial. A study reported that mTORC2 is not required for AKT activation in muscle by genetic ablation of rictor (an essential protein for mTORC2) [76]. In contrast, another report suggested that rictor knock-down in skeletal muscle compromises AKT signaling in this tissue [51].

mTORC1 is one of the most important effectors in this signaling pathway after AKT and is sensitive to rapamycin [66]. Although AKT phosphorylates mTORC1 directly, this phosphorylation does not affect mTORC1 activity [77]. It is known that AKT activates mTORC1 indirectly by phosphorylating and inhibiting Tuberous Sclerosis Complex 2 (TSC2) [78–80]. TSC2 is a protein that forms a complex with Tuberous Sclerosis Complex 1 (TSC1), and this complex serves to inactivate Ras homolog enriched in brain (Rheb), which triggers mTORC1 [81, 82]. Thus, AKT has the capacity to block an inhibitor of mTORC1.

The mTORC1 complex regulates protein synthesis by phosphorylation of the S6 kinases (S6K1 and S6K2). These two kinases phosphorylate the ribosomal protein S6 and other factors involved in the initiation and elongation steps of mRNA translation, and this action stimulates protein synthesis [66, 83]. In fact, knock-out mice for S6K1 display a reduction in myofiber size [84]. Moreover, mTORC1 also activates Eukaryotic translation initiation factor 4E (eIF4E), one of the factors involved in the initiation of protein synthesis. This factor is inhibited by interaction with 4EBP proteins. Thus, mTORC1 phosphorylates these 4EBPS proteins, thereby disrupting this inhibitory interaction and allowing eIF4E to perform its regular function [66, 83].

In addition, mTORC1 is a potent inhibitor of protein degradation. This complex inhibits autophagy by phosphorylating one of the key kinases, Unc-51 like autophagy activating kinase 1 (ULK1), which is involved in the formation of the autophagosome [85]. However, the regulatory role of mTORC1 in autophagy in skeletal muscle is controversial. Thus, some authors have indicated that starvation-induced autophagy in skeletal muscle is not compromised by mTORC1 inhibition with rapamycin or by mTOR knockdown [51]. These results suggest that autophagy is independent of mTORC1 in this tissue [51]. However, others have shown that prolonged activation of mTORC1 in skeletal muscle is characterized by an inhibition of starvation-induced autophagy, together with the development of a myopathy caused by impaired autophagy [86]. Although autophagy activation promotes muscle wasting, basal autophagy is a quality-control mechanism that is crucial for the maintenance of muscle quality, mass, and function [87].

Apart from its action on mTORC1, AKT can also activate protein synthesis by phosphorylating and inhibiting the serine/threonine kinase GSK3β. In addition to its role in regulating glucose metabolism, GSK3β can also inhibit protein synthesis. More specifically, it interacts with Eukaryotic translation initiation factor 2B (eIF2B), one of the factors that participates in the initiation of protein translation [83, 88, 89]. Thus, when GSK3β is phosphorylated by AKT, this interaction is disrupted and protein synthesis is activated [83, 88, 89].

Finally, other targets of AKT are the forkhead box O (FoxO) transcription factors. Three members of this family are expressed in skeletal muscle (FoxO1, FoxO3, and FoxO4) and have been reported to be major regulators of the atrophy program. One of the initial lines of evidence supporting this role is that the overexpression of FoxO1 in skeletal muscle causes muscle atrophy [90]. Later on, it was also reported that transfection of FoxO3 in skeletal muscle in vivo is sufficient to promote protein degradation. More specifically, FoxO3 induces the expression of several genes involved in protein degradation via the UPS, including two muscle-specific E3 ubiquitin ligases (MuRF1 and atrogin 1), and various proteasome subunits [48, 49, 52, 89]. Moreover, FoxO3 not only enhances UPS activity in skeletal muscle but also induces autophagy in this tissue [51]. Thus, while overexpression of a constitutively active form of FoxO3 promotes autophagosome formation in skeletal muscle, FoxO3 inhibition by overexpression of a dominant negative form or by interference RNA inhibits starvation-induced autophagosome formation [51]. In fact, FoxO3 is required for the induction of several genes involved in autophagy, namely LC3, Bnip3, Atg12, Beclin1, Atg4b, ULK1, Vps34, Bnip3l, and GABARAPL1 [51]. These observations indicate that FoxO3 is one of the major regulators of skeletal muscle mass, as it coordinately regulates the UPS and autophagy in this tissue [48, 51].

AKT is a negative regulator of FoxO transcription factor. FoxO phosphorylation by AKT causes the translocation of this factor from the nucleus to the cytosol, where it is sequestered by the 14-3-3 chaperones [49, 91], thus inhibiting its function as an activator of transcription. Moreover, in extreme cases where AKT stimulation becomes chronic, this phosphorylation promotes the degradation of FoxO transcription factors by the UPS [92] (Fig. 1).

Protein metabolism and muscle mass in type 1 diabetes mellitus

During the 20th century, type 1 diabetes mellitus was extensively studied and its effects on whole-body metabolism were well characterized. Thus, withdrawal of insulin treatment in type 1 diabetic subjects causes a highly catabolic state characterized by an increased protein degradation rate that produces an accelerated loss of muscle mass, together with a high negative nitrogen balance [20–23, 25] (Table 1).

Table 1.

Comparison between human and murine diabetes in terms of muscle mass

	Type 1	Type 2
Diabetes mellitus (human)
Skeletal muscle mass	Muscle atrophy	Muscle atrophy	No changes or increase
Murine diabetes
Most common models	STZ injection	db/db mouse	HFD
		ob/ob mouse
		ZDF rat
		GK rat
		POUND mouse
Skeletal muscle mass	Muscle atrophy	Muscle atrophy	No changes

Although whole-body protein degradation is increased in type 1 diabetes mainly due to the increase in protein degradation in skeletal muscle, it is noteworthy that protein synthesis increases in the splanchnic region of patients [23]. This observation could be explained by the increase in the availability of amino acids as a result of the enhanced protein degradation in skeletal muscle. Both protein synthesis and degradation require a large amount of energy, thus leading to an increase in energy expenditure [24]. The increase in this expenditure in type 1 diabetes may also be attributable to high circulating levels of glucagon and enhanced hepatic gluconeogenesis [93, 94].

Type 1 diabetes in murine models is classically studied by the induction of the disease by streptozotocin injection, which causes a rapid increase in blood glucose levels and induces muscle wasting [57]. These manifestations are very similar to those displayed by type 1 diabetic patients upon withdrawal of insulin treatment. Studies performed in insulinopenic murine models show an increase in UPS activity in skeletal muscle [46, 50, 95, 96]. Moreover, in streptozotocin-induced diabetes, there is an activation of autophagy, which also participates in the loss of muscle mass [57].

Insulin deficiency has been considered to be the main cause of muscle atrophy in subjects with type 1 diabetes, as insulin treatment prevents muscle loss. However, studies performed in murine models show that alterations in other signaling pathways can also contribute to this loss. First, circulating IGF-1 levels are reduced in type 1 diabetes patients [97, 98]. IGF-1 is an anabolic factor for skeletal muscle and is involved in the regulation of skeletal muscle mass [99]. More specifically, IGF-1 overexpression enhances muscle mass and strength [100], while muscle-specific IGF-1 receptor ablation impairs muscle development and causes muscle atrophy [101]. IGF-1 shares the same signaling pathway as that used by insulin.

Another factor to take into consideration is cortisol. Circulating levels of this hormone are often increased in type 1 diabetic subjects [102]. Glucocorticoids have been tightly linked to an increase in protein degradation in skeletal muscle in several conditions, such as fasting, acidosis, and insulinopenic diabetes [39, 103, 104]. From a mechanistic standpoint, glucocorticoids activate the UPS by increasing the expression of several subunits of the proteasome [39, 103, 104].

Finally, IL-6 is also elevated in several type 1 diabetic subjects, especially children [105–108]. In fact, IL-6 is increased in several conditions associated with muscle wasting and it promotes muscle loss [109–112]. The main intracellular mediator of IL-6 signaling inside the cells is STAT3 [113, 114]. In accordance, STAT3 is chronically activated in myofibers in several muscle-wasting conditions, such as cancer cachexia and chronic kidney disease [109–112]. The activation of IL-6–STAT3 signaling in skeletal muscle under these pathological conditions induces myostatin expression, a negative regulator of muscle mass [109, 115–117]. In this regard, specific STAT3 ablation in skeletal muscle ameliorates muscle wasting in a mouse model of streptozotocin-induced diabetes [109].

Thus, the relationship between the signaling pathways that are altered in insulinopenic diabetes is highly complex and still not fully understood (Fig. 2).

Fig. 2.

Relevant signaling pathways involved in skeletal muscle mass regulation. a Muscle mass homeostasis is maintained by a balance between protein synthesis and degradation. Insulin and IGF-1 promote protein synthesis in skeletal muscle. Moreover, insulin reduces protein degradation by inhibiting both the UPS and autophagy. On the other hand, cortisol and myostatin stimulate protein degradation. Cortisol enhances UPS activity, while myostatin activates both the UPS and autophagy. b Alterations found in insulinopenic diabetes that leads to muscle atrophy (reviewed in [21]). The absence of insulin and the reduced circulating levels of IGF-1 compromise protein synthesis. On the other hand, protein degradation is enhanced by the lack of insulin and the increase in the levels of cortisol and myostatin

Protein metabolism and muscle mass in type 2 diabetes mellitus and insulin resistance

In contrast to what occurs in type 1 diabetes, muscle loss is not a clinical feature of most insulin-resistant, obese, or type 2 diabetic subjects (Table 1). Thus, no changes in muscle mass or even increased muscle weight are detected in obese subjects [26], and some studies have demonstrated a positive correlation between skeletal muscle mass and body fat content [27, 28] (Table 1). In fact, a number of authors suggest that type 2 diabetes is associated with muscle loss only in specific clinical contexts. One important factor is the appearance of complications associated with type 2 diabetes, such as cardiovascular comorbidity and chronic kidney disease [118–120]. These conditions have been associated with increased IL-6 and myostatin levels—factors that favor muscle atrophy, as previously commented [109, 121, 122].

Another factor is age, as type 2 diabetes is a risk factor for sarcopenia [29, 123, 124]. According to the Korean Sarcopenic Obesity Study, elderly subjects with diabetes have three times more risk of developing sarcopenia than healthy ones [124]. In this study, both non-diabetic and type 2 diabetic subjects had similar BMIs [124]. Sarcopenia has also been associated with obesity and insulin resistance, a condition named sarcopenic obesity [125, 126]. More specifically, the presence of both obesity and insulin resistance in sarcopenic patients enhances the impairment in skeletal muscle function already present in sarcopenia [125, 126]. Given the changes in endocrine function caused by obesity, a reduction in the circulating levels of factors that promote protein synthesis in skeletal muscle, such as testosterone, adiponectin or IGF-1, together with an increase in factors that enhance protein degradation in this tissue, such as TNFα, IL-6, leptin and myostatin, may be involved in this process (reviewed in [127]). However, the question as to the nature of the mechanisms involved in sarcopenic and why only a minority of patients develop this condition remains unclear.

Regarding the way in which protein metabolism is affected in type 2 diabetes, controversial results have been reported. Various authors indicate that type 2 diabetic subjects adapt to the high levels of circulating insulin, thus maintaining the same level of protein degradation as healthy individuals [128, 129]. These authors have also reported that the anabolic response of skeletal muscle protein metabolism to insulin infusion is very similar in diabetic and healthy subjects [128, 129]. On the other hand, other authors have described that insulin resistance is involved in protein metabolism [130, 131]. More specifically, they have shown that, upon insulin infusion, type 2 diabetic subjects do not display increased protein anabolism or inhibition of protein catabolism. Some authors argue that the use of different approaches and selection criteria for the individuals included in the studies may account for these discrepancies [130, 132, 133]. Proteomic analysis has documented increased expression of proteasomal subunits in muscles from type 2 diabetic patients, and surprisingly dysregulated expression of some proteasome subunits in response to insulin has been reported in myotubes derived from type 2 diabetic patients [134, 135].

Overall, current knowledge indicates that in basal conditions, type 2 diabetic subjects do not show any alteration in whole-protein turnover or net loss of skeletal muscle protein compared to healthy controls. This observation might seem paradoxical given that insulin inhibits protein degradation in human skeletal muscle [23, 25, 53–55]. This finding suggests that there is a mechanism to preserve muscle mass under these pathological conditions. Little is known about the factors and regulators of this mechanism. One of the proteins proposed to be involved in this adaptive response is TP53INP2. This molecule promotes autophagy and muscle wasting in skeletal muscle in murine models but is repressed in type 2 diabetic subjects and in overweight subjects that start to develop insulin resistance [57] (further discussion about TP53INP2 will be given in the following sections). Thus, there are still many open questions regarding this topic: first, the identification and characterization of other factors that could be involved in this mechanism; second, how the expression or activity of these factors is regulated under insulin resistance; and third, why these mechanisms are observed specifically in type 2 and not type 1 diabetes mellitus. One possible explanation is that the long and progressive onset of insulin resistance and type 2 diabetes allows skeletal muscle to adapt to this new condition of impaired insulin function. The onset of type 1 diabetes mellitus is much shorter than for type 2. However, this hypothesis has yet to be demonstrated.

Further research is needed to clearly define this putative adaptive mechanism and its regulation and to establish whether this could have a clinical application for the treatment of muscle wasting. In this regard, it is noteworthy that most murine models of insulin resistance, obesity, and type 2 diabetes do not emulate the clinical features of skeletal muscle mass preservation seen in most patients. This point will be discussed in further detail in the following sections.

A new regulatory protein in autophagy: TP53INP2

TP53INP2, also referred to as diabetes and obesity regulated (DOR) gene, is the only homolog of tumor protein 53-interacting protein 1 (TP53INP1) [159], a regulator of p53 and p73 protein activity [160, 161]. Both TP53INP1 and TP53INP2 genes are present in metazoan species [159]. TP53INP2 is abundantly expressed in highly metabolic adult mouse tissues (such as skeletal muscle, heart, or brain) [57, 162].

DOR/TP53INP2 was initially described as a nuclear protein able to transactivate nuclear hormone receptors such as TRα1, GR, PPARγ, and VDR in the presence of the respective ligand and in a dose-dependent manner in mammalian cells [159, 162]. However, TP53INP2 also has a second function as an activator of autophagy [57, 164]. In this review, we will focus on the role of TP53INP2 as an autophagy regulator.

Under basal conditions, TP53INP2 is mainly nuclear, and it constantly shuttles between the nucleus and the cytosol [163, 164]. However, under conditions characterized by the induction of autophagy, TP53INP2 moves to the cytosol and colocalizes with autophagosomes [164, 165]. TP53INP2 is involved in the initial stages of autophagy through direct interaction with the Atg8 proteins, LC3, GABARAP, GABARAPL1, and GATE16 (which are essential for autophagosome formation) [159, 164, 165]. TP53INP2 may operate as an autophagy receptor protein for ubiquitinated proteins. In support of this view, TP53INP2 co-immunoprecipitates with ubiquitinated proteins, most likely by binding to mono and K63-linked ubiquitin [57]. Recent findings also point to the participation of TP53INP2 in the nuclear exit of LC3 to initiate autophagy [166]. TP53INP2 gain-of-function causes enhanced protein degradation and increases the number of autophagosomes in cells under basal conditions or upon amino acid starvation conditions (a stimulus that activates autophagy) [164, 165]. Conversely, loss-of-function of this protein produces a reduced rate of protein degradation and a decrease in the number of autophagosomes both at baseline and in amino acid starvation [164, 165]. These stimulatory effects of TP53INP2 on autophagy have been documented both in mammalian cells and fly cells [164]. A model of the functional role of TP53INP2 in autophagy is shown in Fig. 3. Studies performed in transgenic mice also support the notion that TP53INP2 regulates skeletal muscle autophagy [57]. TP53INP2 gain-of-function in skeletal muscle is characterized by an increase in the LC3II/LC3I ratio and greater accumulation of LC3II protein and of ubiquitinated proteins upon chloroquine-induced lysosomal inhibition [57]. In contrast, skeletal muscle-specific TP53INP2 ablation increases the protein content of both LC3I and LC3II and causes a lower accumulation of LC3II than controls in response to chloroquine [57].

Fig. 3.

Functional role of TP53INP2 in autophagy. TP53INP2 interacts with LC3 in the nucleus. Upon autophagy induction, TP53INP2 translocates from the nucleus to the cytosol together with LC3. In the cytosol, TP53INP2 also interacts with other Atg8 family members such as GATE 16, and promotes autophagosome formation. Upon closure, TP53INP2 leaves the autophagosome before the fusion with the lysosome. Under insulin-resistant conditions, TP53INP2 gene expression is repressed, which permits to downregulate an excessive autophagy

TP53INP2 promotes muscle loss and is regulated in type 2 diabetes

TP53INP2 controls muscle mass in mice. Transgenic lines overexpressing TP53INP2 specifically in skeletal muscle show a reduction in muscle weight and a decrease in myofiber cross-sectional area [57]. In contrast, SKM-KO mice (with specific ablation of TP53INP2 in skeletal muscle) show muscle hypertrophy, together with increased cross-sectional area of muscle fibers [57]. The induction of diabetes by streptozotocin administration causes enhanced muscle loss in mice overexpressing TP53INP2 compared to wild-type animals, while SKM-KO mice treated with streptozotocin lose less muscle mass than control littermates [57]. The effects of TP53INP2 on muscle mass depend on its role as autophagy activator [57]. Overall, available data support the notion that autophagy preserves skeletal muscle mass and quality [87]. In this regard, autophagy blockage causes the accumulation of damaged and abnormal organelles, which alter muscle structure and function, thus leading to muscle atrophy [87]. However, modulation above a certain threshold also causes muscle atrophy by excessive protein degradation [57].

TP53INP2 is subjected to regulation in skeletal muscle. Rodent models of diabetes such as ZDF rats, db/db mice, and streptozotocin-treated mice display a reduction in muscle TP53INP2 protein levels [57]. Given that most of these models are characterized by muscle atrophy [46, 50, 95, 96] and that TP53INP2 is a negative regulator of muscle mass, the repression of this protein may represent a mechanism by which muscle mass is spared under catabolic insulinopenic conditions.

As previously mentioned, most subjects with type 2 diabetes do not display muscle loss in spite of being insulin-resistant [26–28]. This observation supports the notion of an adaptive mechanism that protects skeletal muscle from accelerated muscle loss under this pathological condition. Muscle TP53INP2 gene expression is lower both in type 2 diabetic and in obese non-diabetic individuals compared to non-obese control subjects [57]. In addition, TP53INP2 mRNA levels are lower in skeletal muscle of overweight insulin-resistant subjects compared to lean subjects, thereby indicating that in humans insulin resistance per se causes TP53INP2 repression in this tissue [57]. Overall, these data indicate that TP53INP2 repression may be part of the mechanism that prevents muscle loss upon deficient insulin signaling in skeletal muscle [57]. The mechanisms responsible for the repressed expression of TP53INP2 under insulin resistance are presently unknown.

Rodent models of type 2 diabetes mellitus

Several rat and mouse models are used as paradigms of type 2 diabetes mellitus. These animals reproduce some of the alterations in glucose metabolism and lipid metabolism found in humans. Many animal models of type 2 diabetes are obese, reflecting the human condition, in which obesity is closely linked to the development of this pathology. The most widely used monogenic models of obesity and diabetes are caused by mutations in genes encoding proteins involved in leptin signaling, leading to hyperphagia and subsequent obesity. These models include the ob/ob mouse (or Lep^ob/ob), which is deficient in leptin, and the db/db mouse (or Lepr^db/db) and Zucker Diabetic Fatty (ZDF) rat, which are deficient in the leptin receptor.

The ob/ob mouse is a model of severe obesity. This paradigm originated in a C57BL/6J genetic background at the Jackson Laboratory in 1949, and it is caused by mutations in the leptin gene [136]. The weight increase starts at 2 weeks of age, and the mice develop hyperinsulinemia. By 4 weeks, hyperglycemia is apparent and it peaks at 3–5 months [137]. Other metabolic alterations include hyperlipidemia, dysregulated body temperature, and reduced physical activity [137]. The pancreatic β-cell mass is dramatically increased in ob/ob mice, and insulin secretion is maintained [138, 139], thus the resulting diabetes is not severe. In contrast, ob/ob mice in the C57BL/Ks background develop a more severe type of diabetes characterized by the regression of islets [140].

The db/db mouse, which is caused by an autosomal recessive mutation in the leptin receptor [141], also originated from the Jackson Laboratory [142]. Mice are hyperphagic, obese, hyperinsulinemic, and hyperglycemic. Obesity is detected from 3 to 4 weeks of age with hyperinsulinemia becoming apparent at around 2 weeks of age and hyperglycemia at 4–8 weeks.

Zucker Fatty rats were discovered in 1961 after crossing the Merck M-strain and Sherman rats. This diabetes model has a mutated leptin receptor [143] that induces hyperphagia, and the rats become obese at around 4 weeks of age [143]. These animals are hyperinsulinemic, hyperlipidemic, and hypertensive, and they show impaired glucose tolerance [144]. A mutation in this strain has led to a diabetogenic phenotype, namely the inbred Zucker Diabetic Fatty Rats (ZDF). These rats are less obese than the Zucker fatty rats but have more severe insulin resistance, which they are unable to compensate because of increased β-cell apoptosis [145]. Insulin resistance is characterized by initial hyperinsulinemia at around 8 weeks of age followed by decreased insulin levels [146]. Diabetes usually develops at around 8–10 weeks in males. In contrast, females do not develop overt diabetes.

Alternatively, obesity can be induced by a high-fat diet (by replacing a normal diet containing around 11 % fat for a diet containing around 60 % of fat). The model of high-fat feeding to C57BL/6 mice was first described in 1988 [147]. High-fat feeding can lead to obesity, hyperinsulinemia, and altered glucose homeostasis as a result of insufficient compensation by islets [148]. It has been shown that mice fed high-fat diets can weigh more than chow-fed controls within a week of starting the high-fat regime [148], although typically mice are fed the high-fat diet for several weeks to induce a more pronounced weight gain. The weight gain is associated with insulin resistance, and lack of β-cell compensation leads to impaired glucose tolerance.

Animal models of lean type 2 diabetes have also been generated. Goto–Kakizaki (GK) rats were generated by repetitive breeding of Wistar rats with the poorest glucose tolerance [149]. GK diabetic rats are characterized by glucose intolerance and defective glucose-induced insulin secretion. The development of insulin resistance does not seem to be the main initiator of hyperglycemia in this model, and the defective glucose metabolism is regarded to be due to aberrant β-cell mass and/or function [150, 151]. However, islet morphology and metabolism seem to differ among GK colonies, and whereas some colonies show a normal β-cell mass and defective insulin secretion, others show a reduced β-cell mass [150, 151].

Most rodent models of type 2 diabetes present muscle atrophy (Table 1). Thus, ob/ob mice exhibit a reduced skeletal muscle mass and reduced muscle fiber size [152, 153], thus indicating muscle atrophy. Under these conditions, muscle from ob/ob mice shows enhanced UPS activity [154]. These effects seem to be specific of defective leptin signaling, and chronic treatment with leptin ameliorates muscle atrophy in ob/ob mice [153]. Furthermore, diabetic db/db mice also show a reduced muscle mass, decreased grip strength, and greater proteasome activity [120, 155]. In addition, enhanced muscle expression of myostatin is detected in db/db mice, and in this regard, genetic ablation of myostatin increases muscle mass in these animals, although it does not completely rescue the defects of diabetes [155]. POUND mice, which lack all leptin receptor isoforms, also show reduced muscle mass [156].

Diabetic GK rats also present a reduction in muscle mass and a decreased muscle fiber size [157]. Under these conditions, the expression of autophagic genes and the abundance of autophagosomes are increased in the muscle of these animals [157].

In contrast to most models of type 2 diabetes, obesity induced by a high-fat diet for 14 weeks does not alter muscle mass in mice (Table 1) [158]. Nevertheless, it reduces the hypertrophic response to mechanical overloading in plantaris muscle [158]. In all, available data in rat or mouse models of type 2 diabetes reveal muscle atrophy. These alterations occur independently of obesity and may be a consequence of hyperglycemia since muscle atrophy is not detected in euglycemic high-fat-fed mice. This pattern of changes in animal models of type 2 diabetes clearly differs from the alterations reported in most type 2 diabetic subjects. These comparisons highlight, among other things, the necessity to develop new murine models of diabetes that better mimic the pattern of changes, including those in protein metabolism and muscle mass, detected in human type 2 diabetes.

Future perspectives

The prevention or amelioration of muscle loss that occurs in a number of pathological conditions, including myopathies, cancer cachexia, type 1 diabetes, and aging, is a major unmet medical need. In this respect, it is relevant to identify druggable regulatory factors that modulate protein degradation or protein synthesis in skeletal muscle and that allow an increase in muscle mass and improved muscle function under muscle-wasting conditions.

Specifically, in the topic of the regulation of muscle mass in diabetic states, there is a clear need to identify the mechanisms that promote muscle wasting in type 1 diabetes and to identify the factors that allow muscle maintenance in type 2 diabetes. In this regard, to understand the factors that maintain muscle homeostasis in type 2 diabetes, it would be pertinent to identify additional factors that regulate autophagy or the UPS in skeletal muscle. These factors, in particular those that contribute to preserving muscle mass in type 2 diabetes, may be particularly relevant since they may be suitable targets for novel anti-muscle-wasting drugs. Thus, the main challenge will be the development of new murine models of insulin resistance and type 2 diabetes mellitus that better mimic the features observed in the human disease with regard to muscle mass.

Abbreviations

4EBP

Eukaryotic translation initiation factor 4E-binding protein

AKT

v-Akt murine thymoma viral oncogene homolog

Atg4b

Autophagy-related protein 4b

Atg8

Autophagy-related protein 8

Atg12

Autophagy-related protein 12

ATP

Adenosine triphosphate

BMI

Body mass index

Bnip3

BCL2/adenovirus E1B 19-kDa protein-interacting protein 3

Bnip3l

BCL2/adenovirus E1B 19-kDa protein-interacting protein 3-like

DOR

Diabetes and obesity regulated

EDL

Extensor digitorum longus

eIF2B

Eukaryotic translation initiation factor 2B

eIF4E

Eukaryotic translation initiation factor 4E

FoxO

Forkhead box O

GABARAP

Gamma-aminobutyric acid receptor-associated protein

GABARAPL1

Gamma-aminobutyric acid receptor-associated protein-like 1

GATE16

Golgi-associated ATPase enhancer of 16 kDa

GR

Glucocorticoid receptor

GSK3β

Glycogen synthase kinase 3 beta

GTP

Guanosine triphosphate

IGF-1

Insulin-like growth factor-1

IL-6

Interleukin-6

IRS

Insulin receptor substrate

LC3

Microtubule-associated protein 1 light chain 3

Lep

Leptin

Lepr

Leptin receptor

MuRF1

Muscle RING finger 1

mTOR

Mammalian target of rapamycin

mTORC1

Mammalian target of rapamycin complex 1

mTORC2

Mammalian target of rapamycin complex 2

PDK1

3-Phosphoinositide-dependent protein kinase-1

PI3K

Phosphatidylinositol 3-kinase

PML

Promyelocytic leukemia

PPARγ

Peroxisome proliferator-activated receptor gamma

Rheb

Ras homolog enriched in brain

STAT3

Signal transducer and activator of transcription 3

S6

Ribosomal protein S6

S6K1

Ribosomal protein S6 kinase 1

S6K2

Ribosomal protein S6 kinase 2

TNFα

Tumor necrosis factor α

TP53INP1

Tumor protein p53-inducible nuclear protein 1

TP53INP2

Tumor protein p53-inducible nuclear protein 2

TRα1

Thyroid hormone receptor alpha large isoform

TSC1

Tuberous sclerosis complex 1

TSC2

Tuberous sclerosis complex 2

ULK1

Unc-51-like autophagy-activating kinase 1

UPS

Ubiquitin proteasome system

Vps34

Phosphatidylinositol 3-kinase Vps34

VDR

Vitamin D3 receptor