Sarcopenia Age-Related Muscle Wasting and Weakness: Mechanisms and Treatments P29 potx

Therefore, for both technical and biological reasons, the current mass spectrometric recording of the electrophoretically or chromatographically separated muscle protein complement can o

Name

Isolectric point

Molecular mass

Peptides matched

Mascot score

+ -transporting

RGYLDKLEPSKI KTSIAIDTIINQKR KGIRP

(Piec et al 2005; Gelfi et al 2006a; O’Connell et al 2007; Doran et al 2008; Feng

et al 2008; Lombardi et al 2009) It is, however, important to stress that motor neurons form an integral part of the physiological units that regulate and maintain excitation–contraction coupling and muscle relaxation In contrast to the relatively stable skeletal muscle genome, the fibre proteome does not exist as a distinct cohort

of biomolecules For obvious biological reasons, any tissue-specific protein com-plement is constantly changing and adapting to altered physiological and pathologi-cal demands This phenomena is even more pronounced in the case of the muscle proteome, since skeletal muscles belong to the class of excessively plastic and adaptable tissues (Pette 2001; Flueck and Hoppeler 2003) The heterogeneous char-acter of individual muscles and the inescapable influence of neuromuscular activity

on fibre distribution make the proteomic profiling of diseased or aged muscles more complex as compared to many other tissues Besides biological considerations, another major hurdle for the comprehensive cataloging and differential analysis of muscle proteomes is the concentration range of proteins It is currently difficult to accurately determine differences in skeletal muscle protein density

However, proteomic studies have determined the dynamic range of plasma pro-tein concentrations and predict that at least nine orders of magnitude separate one of the most abundant elements of this body fluid, albumin, and the rarest protein in this body fluid, interleukin-6 (Pieper et al 2003) The concentration range of plasma proteins involved in immune defense, coagulation and metabolite transportation has been estimated from pg/ml-values at the low abundance end to mg/ml-values at the high abundance end (Anderson and Anderson 2002) A similar dynamic range in protein concentration probably also exists in contractile tissues If one takes into account the fact that the human genome consists of approximate 30,000 genes which

in turn produce several 100,000 individual proteins, it is safe to assume that the number of protein isoforms in the skeletal muscle proteome exceeds the number of muscle-specific genes Therefore, for both technical and biological reasons, the current mass spectrometric recording of the electrophoretically or chromatographically separated muscle protein complement can only represent a partial documentation of the entire fibre proteome Even the most sophisticated approaches for the simultane-ous visualization of the soluble components derived from a specific proteome, such

as fluorescence difference in-gel electrophoresis (Viswanathan et al 2006), can only separate a few thousand proteins (Doran et al 2006) Thus, even proteomic studies

of tissues with a relatively low number of individual classes of proteins and a con-siderably narrower range of protein concentrations as observed in plasma, can only determine the near-to-total proteome

Over the last few years, muscle proteomics has identified the most abundant components of contractile fibres from various species, including humans and the most important animal species used for biomedical research Most studies have focused on the total soluble protein complement, but more discriminatory approaches covering low-abundance elements from distinct subcellular fractions and membrane-associated proteins are emerging The cataloguing of total muscle proteomes has included tissues derived from mouse (Raddatz et al 2008), rat (Yan et al 2001), rabbit (Donoghue et al 2007), chicken (Doherty et al 2004),

sheep (Hamelin et al 2007), pig (Kim et al 2004), cow (Bouley et al 2005) and human (Gelfi et al 2003) Subproteomic profiles have been reported for the cyto-solic, microsomal, nuclear and mitochondrial fraction (Forner et al 2006; Vitorino et al 2007) Muscle protein expression levels were determined under developmental, physiological, pathological and aging conditions Comparative studies have included the proteomic characterization of myoblast differentiation (Kislinger et al 2005), muscle transformation (Donoghue et al 2005), the effect

of endurance exercise (Burniston 2008), muscular hypertrophy (Hamelin et al

(Vigano et al 2008), sepsis-related muscle damage (Duan et al 2006), hypoxia-associated metabolic modulations (De Palma et al 2007), neonatal muscle fibre necrosis of postural muscles (Le Bihan et al 2006), denervation–reinnervation cycles (Sun et al 2006), x-linked muscular dystrophy (Doran et al 2006), dysfer-lionpathy (De Palma et al 2006) and aging (Doran et al 2008) Post mortem changes in the fibre proteome have been profiled for agriculturally important animal muscles, i.e bovine and porcine meat (Lametsch and Bendixen 2001; Jia

et al 2006) Since post-translational modifications (PTM) play a crucial role in protein function and are responsible for much of the heterogeneity in muscle proteins, the establishment of proteomic maps based on common PTMs has been initiated This includes the identification of critical glycosylation, phosphoryla-tion, nitration and carbonylation sites and their role in health and disease (Kanski

et al 2005; Meany et al 2007; Gannon et al 2008; O’Connell et al 2008a; Feng

et al 2008)

3 Proteomics of Muscle Aging

To better understand aging of the neuromuscular system, numerous proteomic studies have been carried out over the last few years Major studies are listed in Table 2 The usual workflow of gel electrophoresis-based proteomic studies of muscle aging and the subsequent biochemical and cell biological characterization

of novel mass spectrometry-identified protein markers is illustrated in Fig 2 The general trend of altered protein expression patterns agrees with the findings from previous physiological, biochemical, cell biological and genomic studies (Piec

et al 2005; Gelfi et al 2006a; Dencher et al 2006, 2007; O’Connell et al 2007; Doran et al 2007c, 2008; Lombardi et al 2009; Capitanio et al 2009) However, certain results from transcriptomic analyses of muscle aging do not concur with proteomic investigations (Welle et al 2001; Giresi et al 2005; Dennis et al 2008) Several genes that encode mitochondrial enzymes are down-regulated in aged fibres (Kayo et al 2001), while the protein ratio between mitochondrial and glyco-lytic muscle proteins was shown to be increased (Piec et al 2005; Gelfi et al 2006a; Doran et al 2008) Possibly, age-related alterations at the transcriptional and proteomic level do not correspond for all classes of proteins Transcriptomic investigations have demonstrated that the age-related up-regulation of genes

includes factors involved in stress response, apoptosis, inflammation, proteolysis and neuronal regulation (Roth et al 2002) In contrast, aging is associated with a down-regulation of genes that encode muscle proteins engaged in fibre remodel-ing, the regulation of energy metabolism and muscle growth (Dennis et al 2008) These results indicate that progressive muscle weakness in the elderly is a highly complex process The large-scale proteomic profiling of aging muscles might throw new light on the multi-factorial etiology of sarcopenia and determine the pathobiochemical hierarchy in the many pathways that lead to contractile dysfunction

Previous biomedical studies have established that the loss in skeletal muscle mass and function during aging is associated with a large variety of molecular and cellular abnormalities (Faulkner et al 2007; Edstrom et al 2007) This includes a shift to a slower-twitching fibre population (Prochniewicz et al 2007), decreased protein synthesis of myofibrillar components (Balagopal et al 1997), disturbed ion handling (Schoneich et al 1999), a blunted stress response (Kayani et al 2008), progressive denervation (Carlson 2004), decreased capillarisation (Degens 1998), excitation–contraction uncoupling (Delbono et al 1995), oxidative stress (Squier and Bigelow 2000), mitochondrial dysfunction (Figueiredo et al 2008), increased susceptibility to apoptosis (Dirks and Leeuwenburgh 2002), a metabolic disequilib-rium (Vandervoort and Symons 2001), progressive decline in energy intake (Roberts 1995), a reduced regenerative potential (Renault et al 2002) and inade-quate levels of essential growth factors and hormones indispensable for the main-tenance of the excitation–contraction–relaxation cycle (Lee et al 2007) Proteomics promises to unearth what primary changes within this complex molecular patho-genesis cause detrimental down-stream alterations The large-scale protein bio-chemical analysis of muscle aging may also elucidate what compensatory adaptation processes, repair mechanisms and stress responses are initiated to limit age-dependent fibre degeneration

Table 2 Proteomic profiling studies of sarcopenia of old age

Mass spectrometry-based proteomic

Skeletal muscle type

or fraction References Profiling of total soluble proteome Human Vastus lateralis Gelfi et al 2006a Profiling of total soluble proteome Rat Gastrocnemius Piec et al 2005

O’Connell et al 2007 Doran et al 2008 Lombardi et al 2009 Profiling of motor unit Rat Sciatic nerve and

gastrocnemius

Capitanio et al 2009 Profiling of small heat shock proteins Rat Gastrocnemius Doran et al 2007c PTM analysis of protein glycosylation Rat Gastrocnemius O’Connell et al 2007 PTM analysis of protein nitration Rat Gastrocnemius Kanski et al 2005 PTM analysis of protein carbonylation Rat Mitochondria Feng et al 2008 PTM analysis of protein phosphorylation Rat Gastrocnemius Gannon et al 2008 Subproteomic analysis Rat Mitochondria Dencher et al 2006,

2007

3.1 Remodeling of the Contractile Apparatus during Aging

Major physiological and cell biological differences exist between type-I, type-IIa and type-IIb fibres Differences in motor neuron size, capillary density, myoglobin content, mitochondrial density and metabolite content closely relate to the biochemi-cal composition of the contractile apparatus (Pette and Staron 1990; Punkt 2002; Spangenburg and Booth 2003) A major interest in muscle aging research is to understand what exact changes on the protein level cause a loss of contractile strength in both slow and fast muscles (Prochniewicz et al 2007) The proteomic analysis of aged muscle has revealed a generally perturbed protein expression pat-tern in senescent muscle (Piec et al 2005; O’Connell et al 2007; Doran et al 2008;

Fig 2 Proteomic workflow for the identification and characterization of novel biomarkers of skeletal muscle aging Shown is the gel-electrophoresis (GE)-based separation of young adult ver-sus senescent muscle extracts Two-dimensional gels were stained with colloidal Coomassie Blue (CCB) dye (O’Connell et al 2007 ) Difference in-gel electrophoresis (DIGE) is routinely used for the fluorescent tagging and separation of muscle proteomes and mass spectrometric technology is usually employed to unequivocally identify proteins that exhibit a changed abundance during fibre aging Potential alterations in the biological activity, oligomeric status, expression level, subcellular localization and/or post-translational modifications of newly identified skeletal muscle proteins are then determined by standard biochemical and cell biological methods

Lombardi et al 2009; Capitanio et al 2009), including many of the proteins belonging

to the contractile apparatus that makes up approximately 50% of the total muscle protein complement The supramolecular protein assemblies forming the thick and thin filaments of the basic contractile units exist in a great variety of fibre type-specific isoforms (Pette and Staron 1990) In the presence of ATP, a highly complex and cyclic coupling process between actin filaments and myosin head structures provides the molecular basis for the sliding of thin filaments past thick filaments causing distinct increments of sarcomere shortening (Gordon et al 2000; Fitts

whereby the troponin complex and tropomyosin strands directly regulate and enable actomyosin interactions for force generation (Swartz et al 2006; Kreutziger et al 2007) In skeletal muscles, a close relationship exists between isoform expression patterns of contractile proteins and metabolic fibre properties (Pette and Staron 2001) Thus, to understand the molecular mechanisms that underlie age-related fibre type shifting, a special interest focuses on potential alterations in the isoforms of myosin, actin, troponin or tropomyosin Myosins consist of a hexameric structure consisting of 2 MHC heavy chains and various MLC light chains (Clark et al 2002; Bozzo et al 2005) A recent study by Capitanio et al (2009) has shown a clear age-dependent transformation process within the pool of myosin heavy chain isoforms, i.e a transition from fast MHC-IIb to MHC-IIa to slow MHC-I in 8-month versus 22-month old rat gastrocnemius muscle This pattern of MHC changes is in line with observed adaptive processes in chronic electro-stimulated fast muscle (Pette 2001) and exercised muscles (Sullivan et al 1995)

Fast-to-slow transformation is evidently associated with a shift to more oxidative metabolism and a concomitant change in the aged contractile apparatus to slower kinetics The proteomic profiling of fast muscles following chronic low-frequency stimulation has shown that light and heavy chains of myosin undergo a stepwise replacement from fast to slow isoforms (Donoghue et al 2005, 2007) Previous biochemical studies have shown similar effects of the neuromuscular activity on the expression of individual subunits of troponin (Pette and Staron 2001) In analogy,

a comparable process appears to occur during muscle aging causing a drastic increase in the abundance of slow isoforms of key contractile elements in senescent fibres (Gelfi et al 2006a; Doran et al 2008; Capitanio et al 2009) A comprehen-sive proteomic study of rat muscle aging, using the highly discriminatory fluores-cent difference in-gel electrophoresis technique, has identified the slow myosin light chain isoform MLC-2 as one of the most drastically altered muscle proteins in this animal model of sarcopenia (Doran et al 2008) Thus, both myosin light chains and heavy chains seem to shift towards slower isoforms Application of the phos-pho-specific fluorescent dye ProQ-Diamond demonstrated that the abundance of the slow MLC-2 protein is not only drastically increased, but that its phosphoryla-tion levels are even more enhanced in senescent gastrocnemius fibres (Gannon

et al 2008) This supports the idea of an age-related shift to a slower-twitching fibre population and suggests changed expression levels and altered post-translational modifications in myosin components as novel candidates for establishing a biomarker signature of muscle aging

3.2 Metabolic Adaptations in Aged Skeletal Muscle

Findings from the proteomic analysis of bioenergetic adaptations in aged muscle agree with previous physiological and biochemical studies of fibre aging The results from different proteomic studies of muscle aging have demonstrated that a general shift occurs in major metabolic pathways towards a more oxidative muscle metabolism (Doran et al 2009a) However, species-specific differences appear to exist with respect to the degree of modifications in distinct rate-limiting enzymes and metabolite transporters, as well as in the complexity of these changes in par-ticular pathways (Piec et al 2005; Gelfi et al 2006a; Doran et al 2008) When studying the effects of physiological or pathological factors on contractile function,

it is crucial to take into account the influence of patterns of innervation and activity

on the metabolic and bioenergetic properties of skeletal muscles In the case of diseased and aged muscles, it has clearly been documented that long-term inactivity inevitably results in disuse atrophy which results in a drastic reduction in tissue mass and contractile strength (Kandarian and Jackman 2006) Proteomic studies have to take into account the heterogeneity of skeletal muscles and build on the previous biochemical and physiological knowledge on fibre type characteristics and how they relate to specific marker proteins Distinct protein expression signatures can be conveniently employed to differentiate between type I and type II fibres The abundance and or isform expression pattern of many metabolic enzymes, excita-tion–contraction coupling elements, ion-handling proteins and contractile compo-nents can be used to determine fibre type distributions The proteomic profiling of fast-twitching fibres agrees with a predominantly glycolytic metabolism, a high recruitment frequency, an easily fatigable phenotype and a high maximum power output (Okumura et al 2005; Gelfi et al 2006b) On the other hand, the protein complement of slower fibres is perfectly adapted to oxidative metabolism, a low recruitment frequency, resistance to fatigue and a low maximum power output (Okumura et al 2005; Gelfi et al 2006b) Especially striking is the difference in the density of myosin isoforms, glycolytic enzymes, citric acid cycle enzymes, oxida-tive phosphorylation elements, the oxygen carrier myoglobin and the fatty acid binding protein FABP In addition, the abundance of Ca2+-dependent binding pro-teins, pumps, channels and exchangers differs considerably between fast and slow muscles (Froemming et al 2000) These established fibre type-specific markers could now be used for the interpretation of proteomic profiles generated by mass spectrometry-based muscle aging studies

Major age-dependent alterations in the expression of catabolic enzymes and rate-limiting transporter molecules have been demonstrated by proteomics (Piec

et al 2005; Doran et al 2008) As an example, Fig 3 illustrates the age-related increase in the enzyme adenylate kinase The expression of the soluble AK1 iso-form was shown to be increased using both fluorescent difference in-gel electro-phoresis and two-dimensional immunoblotting Adenylate kinase, in conjunction with creatine kinase, maintaines a major nucleotide pathway in skeletal muscle Increased levels of the AK1 isoform suggest adaptive processes that regulate

nucleotide ratios in aging fibres Other skeletal muscle proteins that exhibit an age-related change in concentration are involved in the transportation of oxygen, the provision of fatty acids and the removal of carbon dioxide, as well as the maintenance of glycolysis, the citric acid cycle and oxidative phosphorylation (Piec et al 2005; Gelfi et al 2006a; Doran et al 2008) Muscle aging is associated with a reduced glycolytic flux due to a drastic reduction in key glycolytic enzymes, such as pyruvate kinase, phosphofructokinase and enolase The reduction of the key regulatory enzyme pyruvate kinase was shown by Deep Purple staining (O’Connell et al 2007), a DIGE-based study (Doran et al 2008) and PTM analysis (O’Connell et al 2008a) Pyruvate kinase facilitates the final oxidoreduction-phosphorylation reaction during glycolysis that converts phosphoenolpyruvate to

29.7

pH Adult muscle

a

b

kDa

CA3

AK - Cy3

AK1

AK1

Adult

Aged

CA3

Aged muscle pH

22.8

21.6

Fig 3 Proteomic profiling of adenylate kinase isoform AK1 in senescent skeletal muscle Shown

is an expanded view of fluorescently tagged two-dimensional gels of the young adult muscle proteome versus the aged muscle proteome Preparations from differently aged rat gastrocnemius

muscles were labelled with the CyDyes Cy3 (a) and Cy5 (b) In panels (c) and (d) are shown the

comparative graphic representation of the AK1 spot in young adult versus aged fibres, respec-tively A major two-dimensional protein spot of approximately 30 kDa represents the abundant muscle enzyme carbonic anhydrase (CA3) The portion of the two-dimensional gel illustrated covers the range of approximately pH 7 to pH 8 in the first dimension and a molecular mass range

of approximately 20–30 kDa in the second dimension While the CA3 spot exhibits comparable levels between adult and aged muscle, the AK1 protein is clearly increased in aged muscle The elevated expression level of adenylate kinase was confirmed by two-dimensional immunoblot (IB)

analysis (e) Standard methods were employed for fluorescent difference in-gel electrophoresis

and immunoblotting (Doran et al 2006 )

ATP and pyruvate (Munoz and Ponce 2003) The decreased expression of the PK-M1 isoform of pyruvate kinase agrees with a shift to more aerobic-oxidative metabolism in senescent muscle Although pyruvate kinase levels are reduced during aging, the remaining cohort of this glycolytic enzyme exhibits drastically increased levels of both N-glycosylation (O’Connell et al 2008a) and tyrosine nitration (Kanski et al 2005) Abnormal post-translational modifications in meta-bolic enzymes are believed to negatively affect the biological activity of glycolytic enzymes, which was shown to be true in the case of the PK-M1 isoform Senescent muscle are characterized by a reduced pyruvate kinase activity (O’Connell et al 2008a) Enhanced N-glycosylation probably influences protein stability, cellular targeting, inter- and intra-molecular interactions, and coupling efficiency between substrates and active site of this enzyme, causing a diminished glycolytic flux rate

in aged fibres In addition, the expression of pyruvate dehydrogenase, the metabolic linker between glycolysis and the citric acid cycle, is lower in aged fibres (Doran

et al 2008) Consequently, the transformation of pyruvate into acetyl-CoA is reduced in sarcopenia The proteomic analysis of the phosphoprotein cohort of aged muscle showed increased phosphorylation for lactate dehydrogenase, albumin and aconitase, and decreased phosphorylation in cytochrome-c-oxidase, creatine kinase and enolase (Gannon et al 2008) Hence, age-related changes in the muscle phosphoproteome are associated with metabolic enzymes from the cytosolic and mitochondrial compartment This agrees with the idea that sarcopenia is a highly complex muscle disease that causes drastic alterations in the expression and molec-ular structure of important metabolic regulators

The biochemical analysis of the fast-to-slow transformation process in chronic electro-stimulated fast muscles strongly suggests that the two most crucial limiting factors of oxidative metabolism are represented by the availability of oxygen and the rate of fatty acid transportation (Kaufmann et al 1989) Since in senescent muscles an up-regulation of both the fatty acid transporter FABP and the oxygen-carrier myoglobin has been demonstrated by proteomic analysis (Doran et al 2008), these alterations in biomarkers suggest that senescent fibres switch to a more aerobic-oxidative metabolism In agreement with this major metabolic adaptation

is the increased expression of citric acid cycle enzymes such as succinate dehydro-genase, isocitrate dehydrogenase and malate dehydrogenase in senescent muscles (Piec et al 2005; Gelfi et al 2006a; Doran et al 2008) Recently Lombardi et al (2009) have determined both the transcriptomic and proteomic profile of aged rat muscle employing a combination of DNA array and native blue PAGE technology Aging seems to differentially affect the abundance, supramolecular organization and activity of the various mitochondrial complexes associated with the oxidative phosphorylation pathway Although aging is generally associated with a shift to more oxidative muscle metabolism, senescent human muscles showed a more pro-nounced transition from predominantly glycolytic to mitochondrial energy genera-tion (Gelfi et al 2006a) as compared to small mammalians such as rats (Piec et al 2005) These species-specific differences should be taken into account in animal model studies The extrapolation of results from aging rat muscle to the human aging process should be undertaken with caution

3.3 Cellular Stress Response in Aged Skeletal Muscle

The fast and efficient up-regulation of stress proteins is an essential cellular survival mechanism that prevents excess protein degradation and deleterious protein aggre-gation during tissue injury (Ellis and van der Vies 1991) In healthy adult muscle fibres, the natural response to stressful conditions involves a diverse array of molecular chaperones, mostly belonging to the very large family of heat shock proteins During fibre adaptation or cellular regeneration phases, molecular chaper-ones stabilize denatured muscle proteins and facilitate the correct folding and con-formational maturation in nascent peptides (McArdle and Jackson 2000) Muscle chaperones protect fibres during extensive contractile activity, traumatic injury, hyperthermia, hypoxic insult, ischemic damage and neuromuscular pathology (Nishimura and Sharp 2005) A common feature of chaperoning heat shock pro-teins is a promotor region that contains a consensus-binding sequence for HSF1 (Amin et al 1988), the heat shock transcription factor that is associated with the response of cells following exposure to acute stressors (Anckar and Sistonen 2007) Heat shock proteins are classified according to their relative molecular mass Besides the widely distributed Hsp60s, Hsp70s, Hsp90s and Hsp100s, some low-molecular-mass heat shock proteins are specifically induced during muscle injury (Golenhofen et al 2004) These small members of the cytoprotective chaperone complement of skeletal muscles are characterized by a a-crystallin domain, a

con-served 90-residue carboxy-terminal sequence (van Montfort et al 2001) A major function of muscle-specific small heat shock proteins is the prevention of deleteri-ous protein aggregation, and they are especially involved in the modulation of intermediate filament assembly (Nicholl and Quinlan 1994)

Heat shock proteins are relatively soluble and abundant, making them ideal can-didates for proteomic investigations Over the last few years, a large number of proteomic studies have identified cellular chaperones in muscle tissues Most mass spectrometry-based analyses showed increased levels of heat shock proteins in the neuromuscular system following exposure to physiological or pathological stressors This included the large-scale screening of fibre transformation following chronic electro-stimulation (Donoghue et al 2005, 2007), moderate intensity endurance exercise (Burniston 2008), myoblast differentiation (Gonnet et al 2008; Tannu et al

denerva-tion (Sun et al 2006), experimental muscular atrophy following hindlimb suspen-sion (Seo et al 2006), dystrophinopathy-associated necrosis (Doran et al 2006), experimental exon-skipping therapy of muscular dystrophy (Doran et al 2009b), dysferlin-related myopathy (De Palma et al 2006), burn sepsis-induced stress (Duan

et al 2006), hypoxia-related stress (Bosworth et al 2005) and post mortem changes

in muscle fibres (Jia et al 2006), as well as age-dependent muscle degeneration (Piec et al 2005; O’Connell et al 2007; Doran et al 2007c; Feng et al 2008; Lombardi et al 2009; Capitanio et al 2009) Interestingly, mass spectrometry-based proteomics of aged muscles has shown increased levels of distinct small chaperones, especially the cardiovascular heat shock protein cvHsp (Doran et al 2007c)