Implications
Skeletal muscle development in different animal species has been an important topic of scientific research for many years. The aim of this review is to provide recent knowledge on the regulation of prenatal muscle growth and its modification by environmental factors in a variety of species, such as pig, sheep, cattle, poultry and fish, and to show the consequences for later growth, carcass quality and meat/flesh quality, which are of great economic importance in animal production.
Introduction
There is increasing evidence that influences on prenatal and early postnatal development of skeletal muscle can result in long-term effects on postnatal growth and physiological function both in animals and in humans. In farm animals, these long-term effects of developmental origin have an impact on economically important traits such as vitality, growth performance, body composition and meat quality. However, the ways in which genotype, genome and physiology contribute to this interaction remains uncertain. Prenatal growth is determined by the genotype of the conceptus, and also depends on the maternal uterine or in ovo milieu of hormones, nutrients and growth factors. Understanding mechanisms that control prenatal development and growth are critical to define strategies to reduce the incidence of developmental disorders and their long-term consequences.
In order to intensify research in this field and to stimulate scientific discussion on the developmental origin of farm animal performance, a European network, COST action 925 ‘The importance of prenatal events for postnatal muscle growth in relation to the quality of muscle based foods’ was established. This European Union-supported COST action (2004 to 2008) brought together scientists from 20 European countries and promoted the mutual exchange of scientific knowledge and resulted in a series of collaborative research projects. Moreover, training of young scientists in this research area was a priority and training grants were available for research visits to participating institutions of the member states.
This review will focus on the main outcomes of research based on the scientific results obtained from experiments with mammals, poultry and fish over 4 years (2004 to 2008) by participants of the COST 925 action. These include prenatal regulatory mechanisms of skeletal muscle development, genetic and environmental impact on muscle development and its consequences for animal performance and meat quality as well as advances and standardisation of research methodologies in this area. Moreover, this review will identify the main gaps and open questions in this field of research. The present part 1 of the review focuses on the regulation of myogenesis and the impact of environment on muscle development and related changes in animal performance (for part 2, see Rehfeldt et al., Reference Rehfeldt, Te Pas, Wimmers, Brameld, Nissen, Berri, Valente, Power, Picard, Stickland and Oksbjerg2011).
Regulatory mechanisms of skeletal muscle development (myogenesis) and growth
Myogenesis and myogenic regulatory factors (MRFs)
The process whereby the skeletal muscle lineage is determined and develops into the mature muscle fibre is termed myogenesis. Evidence to date indicates that the processes are very similar across mammals, poultry and fish, although the timings for the various steps differ considerably. Myogenesis is regulated by a group of muscle-specific transcription factors, collectively known as the muscle regulatory factors or MRFs, and they appear to have the same functions across the various species. There are four MRFs (myf-5, myoD, myogenin and MRF4). Myf-5 and myoD are required for the commitment to the muscle cell type, whereas myogenin and MRF4 are important for inducing differentiation and formation of muscle fibres (see reviews by Te Pas and Soumillion, Reference Te Pas and Soumillion2001; Brameld et al., Reference Brameld, Fahey, Langley-Evans and Buttery2003; Rehfeldt et al., Reference Rehfeldt, Fiedler and Stickland2004a; Stickland et al., Reference Stickland, Bayol, Ashton and Rehfeldt2004; Johnston, Reference Johnston2006; Chang, Reference Chang2007; Brameld and Daniel, Reference Brameld and Daniel2008).
Distinct phases of myofibre formation
There are two or three distinct phases of muscle fibre formation in mammals, poultry and fish, although the terminology used differs. Skeletal trunk muscle development, in mammals and poultry embryo, occurs in two major phases. In the first phase, a primary myotome is formed from cells originating from the four epithelial borders of the dermomyotome. In the second phase, the central region of the dermomyotome undergoes an epithelial to mesenchymal transition and releases pax3/7-positive cells into the primary myotome. These cells evolve into embryonic myoblasts (which give rise to primary fibres) or foetal myoblasts (which give rise to secondary fibres) or satellite cells (which contribute to further growth and repair of the myotome). On the other hand, limb muscles originate from muscle progenitors that delaminate from the ventrolateral border of the dermomyotome. For more details see, for example, the reviews by Biressi et al. (Reference Biressi, Molinaro and Cossu2007), Buckingham and Vincent (Reference Buckingham and Vincent2009) and Relaix and Marcelle (Reference Relaix and Marcelle2009). In mammals and poultry, primary fibres form first, followed by the secondary fibres that develop around these primaries (Maltin et al., Reference Maltin, Delday, Sinclair, Steven and Sneddon2001; Picard et al., Reference Picard, Lefaucheur, Berri and Duclos2002). In large mammals (e.g. sheep and pigs), tertiary fibres have also been described developing between the secondaries (Lefaucheur et al., Reference Lefaucheur, Edom, Ecolan and Butler-Browne1995), although this is hotly debated at times. In mammals, formation of new muscle fibres takes place in utero, with the numbers of muscle fibres within each muscle thought to be set around the time of birth (see reviews by Maltin et al., Reference Maltin, Delday, Sinclair, Steven and Sneddon2001; Picard et al., Reference Picard, Lefaucheur, Berri and Duclos2002; Brameld et al., Reference Brameld, Fahey, Langley-Evans and Buttery2003; Chang, Reference Chang2007; Brameld and Daniel, Reference Brameld and Daniel2008). However, rats and mice are an exception as muscle fibre formation in some muscles continues during the early neonatal period (Rehfeldt et al., Reference Rehfeldt, Fiedler and Wegner1987; Wilson et al., Reference Wilson, Ross and Harris1988). Determination of the total number of fibres within a muscle is technically difficult, particularly in adult muscles of irregular shape from large mammals. It is therefore questionable how much formation of new fibres takes place in postnatal life, but the majority of fibres in most mammals appear to develop in utero (Rehfeldt et al., Reference Rehfeldt, Fiedler and Stickland2004a). Hence, postnatal growth of muscle fibres in mammals mainly involves increases in fibre size (hypertrophy) rather than numbers of fibres. A population of myogenic cells, termed satellite cells, has been identified in muscle of mammals and poultry, particularly during neonatal and postnatal stages of development. In bovine and sheep foetuses, satellite cells are detected adjacent to the myofibres as early as 2 months and 85 days, respectively (Russell and Oteruelo, Reference Russell and Oteruelo1981; Greenwood et al., Reference Greenwood, Slepetis, Hermanson and Bell1999). Satellite cells proliferate and fuse with existing fibres, thereby increasing DNA content and protein synthetic capacity resulting in muscle fibre hypertrophy.
The development of muscle in chicken resembles that of mammals and the total number of muscle fibres (TNF) does not vary after hatching (Smith, Reference Smith1963; Gollnick et al., Reference Gollnick, Parsons, Riedy and Moore1983). The muscle fibres then grow by hypertrophy, both in length and diameter, a phenomenon that involves satellite cells, myogenic precursors present at the periphery of fibres. Satellite cells appear during late chicken embryogenesis between 13 and 16 days in ovo (Feldman and Stockdale, Reference Feldman and Stockdale1992; Hartley et al., Reference Hartley, Bandman and Yablonka-Reuveni1992). Satellite cells are able to multiply and fuse with fibres providing them with new nuclei for protein anabolism and this phenomenon accounts for more than 98% of the final DNA content of chicken muscle (Moss, Reference Moss1968).
Muscle fibre production in fish occurs in three distinct phases (i) embryonic (Rowlerson and Veggetti, Reference Rowlerson and Veggetti2001), (ii) late embryo to early larval stages (stratified hyperplasia) and (iii) larval to adult stages, when new white fibres are formed around larger diameter fibres, giving the muscle a mosaic appearance (mosaic hyperplasia; Rescan, Reference Rescan2008). During the embryonic phase, the formation of the primary myotome starts with the differentiation of a superficial layer of slow muscle fibres (or adaxial cells) from the most medial paraxial mesoderm cells. Subsequently, and shortly after segmentation, the deep fast muscle fibres differentiate from posterior somite cells. Anterior cells of the somite then migrate to the lateral surface of the primary myotome and contribute muscle progenitor cells for primary myotome expansion/growth (reviewed in Stellabotte and Devoto, Reference Stellabotte and Devoto2007). In contrast to mammals and poultry, fish muscle can continue growing throughout life if the food supply is not limiting (Weatherley and Gill, Reference Weatherley and Gill1987). Moreover, throughout the life cycle of fish, both hyperplastic and hypertrophic processes contribute to fish muscle growth, controlled by intrinsic and extrinsic regulatory factors (Carpenè et al., Reference Carpenè, Veggetti and Mascarello1982; Stickland, Reference Stickland1983; Kiessling et al., Reference Kiessling, Johansson and Kiessling1990; Fauconneau et al., Reference Fauconneau, Andre, Chmaitilly, Le Bail, Krieg and Kaushik1997; Campinho et al., Reference Campinho, Sweeney and Power2006; Silva et al., Reference Silva, Valente, Galante, Andrade, Monteiro and Rocha2009a and Reference Silva, Valente, Olmedo, Galante, Monteiro and Rocha2009b). The proliferation of myogenic progenitor cells (stem cells and their progeny) in fish muscle is thought to be the reason why hyperplasia is possible throughout the life cycle (Johnston et al., Reference Johnston, Abercromby, Vieira, Sigursteindóttir, Kristjánsson, Sibthorpe and Skúlason2004). These cells may be analogous to the satellite cells described earlier in the muscle of postnatal mammals and poultry, although their functions appear to be quite different. Mosaic hyperplasia is the main mechanism for expanding white fibre number in the juvenile and adult stages of the majority of species, continuing until approximately 40% of the maximum fish length (Weatherley et al., Reference Weatherley, Gill and Lobo1988). There is evidence that the relative contribution of increases in fibre number and area to total muscle growth in fish is related both to the growth rate and final size attained by each species (Weatherley et al., Reference Weatherley, Gill and Lobo1988).
Critical timings and mechanisms
The timings for the appearance of the different fibre populations vary enormously, but it is becoming clear that the effects of environmental factors on numbers of muscle fibres formed is dependent on the stage of myogenesis at which exposure takes place. Importantly, the critical period for environmental effects appears to be during muscle cell proliferation, immediately before differentiation and fibre formation (Brameld and Daniel, Reference Brameld and Daniel2008). Hence, factors that stimulate myoblast proliferation and/or delay differentiation appear to increase the numbers of muscle fibres, whereas factors that inhibit myoblast proliferation and/or induce early differentiation appear to decrease the numbers of muscle fibres. Some of the discrepancies between studies published to date may relate to differences in the timing of interventions relative to the timing of fibre formation in the species of interest. Indeed, different muscles within the same species may also develop at slightly different times. One possible common factor within all these studies would be changes in energy metabolism and availability for growth processes, particularly cell proliferation. The partitioning of energy to the gravid uterus is complex but involves the indirect action of the growth hormone–insulin like growth factor (GH–IGF) axis (Bauer et al., Reference Bauer, Harding, Bassett, Breier, Oliver, Gallaher, Evans, Woodall and Gluckman1998; Gluckman and Pinal, Reference Gluckman and Pinal2003; Rehfeldt et al., Reference Rehfeldt, Nissen, Kuhn, Vestergaard, Ender and Oksbjerg2004b).
It is widely agreed that one of the main mechanisms for both environmental and genetic effects on muscle fibre development in mammals, poultry and fish is via effects on the GH–IGF axis. The IGFs (IGF-I and IGF-II) are potent mitogens for most cell types, including myoblasts, but they also stimulate muscle cell differentiation by inducing expression of one of the MRFs, myogenin (see Florini et al., Reference Florini, Ewton and Coolican1996; Oksbjerg et al., Reference Oksbjerg, Gondret and Vestergaard2004; Codina et al., Reference Codina, de la Serrana, Sanchez-Gurmaches, Montserrat, Chistyakova, Navarro and Gutierrez2008). Indeed, spontaneous differentiation in cultured muscle cell lines can be inhibited by blocking local IGF-II production/action (see Florini et al., Reference Florini, Ewton and Coolican1996), and the increased number of muscle fibres observed in double-muscled cattle is associated with a delay in both differentiation and local IGF-II expression (Gerrard and Grant, Reference Gerrard and Grant1994), as well as mutations in the myostatin gene (see reviews by Maltin et al., Reference Maltin, Delday, Sinclair, Steven and Sneddon2001; Picard et al., Reference Picard, Lefaucheur, Berri and Duclos2002; Brameld et al., Reference Brameld, Fahey, Langley-Evans and Buttery2003; Brameld and Daniel, Reference Brameld and Daniel2008). However, a number of other hormones, growth factors (e.g. myostatin) and nutrients have also been shown to regulate myoblast proliferation and differentiation directly and therefore are also likely to play a role (see reviews by Maltin et al., Reference Maltin, Delday, Sinclair, Steven and Sneddon2001; Picard et al., Reference Picard, Lefaucheur, Berri and Duclos2002; Brameld et al., Reference Brameld, Fahey, Langley-Evans and Buttery2003; Brameld and Daniel, Reference Brameld and Daniel2008) including antioxidants (Orzechowski et al., Reference Orzechowski, ̡okociejewska, Pawlikowska and Kruszewski2005), soya-derived isoflavones (Mau et al., Reference Mau, Kalbe, Viergutz, Nürnberg and Rehfeldt2008a) and unsaturated fatty acids (Hurley et al., Reference Hurley, Flux, Salter and Brameld2006; see also ‘Environmental impact on skeletal muscle development and consequences for animal growth, carcass and meat quality’).
Skeletal muscle growth, plasticity and function
The biological characteristics of the muscle and the way in which it grows are major factors influencing overall growth capacity, as well as the quality of the farmed product in mammals, poultry and fish (Johnston, Reference Johnston1999 and Reference Johnston2001; Chang, Reference Chang2007). Although growth rate and growth potential are undoubtedly heritable characteristics, growth of all species is dependent upon environmental factors, such as ambient temperature and food availability, both in the determination of growth per se and in the expression of morphological characteristics (Leatherland, Reference Leatherland1994). Growth and function of muscles relate to their fibre-type composition, particularly in terms of contractile speed (slow v. fast) and metabolism (oxidative v. glycolytic). In mammals at least, the majority of primary fibres appear to initially become slow oxidative (type I) muscle fibres, whereas the majority of secondary and tertiary fibres appear to initially become faster fibre types (types IIA, IIX and IIB, see Picard et al., Reference Picard, Lefaucheur, Berri and Duclos2002). Fibre-type transformations occur during neo- and postnatal development in mammals and are regulated by a variety of factors including innervation, exercise and thyroid hormones (THs; Chang, Reference Chang2007). Importantly, faster fibre types (IIX and IIB) tend to have larger diameters, and are therefore important for growth, but are associated with a whiter colour, reduced lipid content and increased toughness, which are important organoleptic properties of meat. Similarly, the distinctive phases of muscle development in fish are underpinned by molecular changes that probably better adapt muscle function to changing physiological requirements during development (Chauvigné et al., Reference Chauvigné, Rallière, Cauty and Rescan2006; Campinho et al., Reference Campinho, Silva, Nowell, Llewellyn, Sweeney and Power2007; Rescan, Reference Rescan2008). The red and white muscle fibres in fish are arranged in discrete anatomical regions (Bone, Reference Bone1978; Mascarello et al., Reference Mascarello, Rowlerson, Radaelli and Veggetti1995; Moutou et al., Reference Moutou, Silva, Mamuris and Power2005; Silva et al., Reference Silva, Rowlerson, Valente, Olmedo, Monteiro and Rocha2008) largely located in the hypaxial and epaxial muscles along the length of the body and represent 40% to 60% of body mass. Moreover, as fish are poikilotherms and do not regulate body temperature, changes in environmental temperature can have a profound impact on the rate of myogenesis, composition of sub-cellular organelles, patterns of gene expression and the number and size distribution of muscle fibres (reviewed by Johnston, Reference Johnston2006). During the embryonic and larval stages, fish muscle plasticity in response to the environment is usually irreversible due to the rapid pace of ontogenetic change with potential consequences for larval survival as well as long-lasting effects on white muscle cellularity, including the duration and intensity of myotube formation in adult stages.
Candidate gene expression related to muscle development and growth
Expression of specific genes involved in myogenesis studied in vitro or in vivo may highlight important aspects of its regulation at the molecular level. Studies on the temporal expression of important genes during myogenesis could also provide insight in defining critical windows prone to nutritional or other manipulations (see ‘Environmental impact on skeletal muscle development and consequences for animal growth, carcass and meat quality’). Future aquaculture and animal production systems may benefit from results enabling monitoring of skeletal muscle development of species.
Using the murine C2C12 satellite cell line Kwiecinska et al. (Reference Kwiecinska, Rosziewicz, ̡okociejewska and Orzechowski2005) showed that MAP kinase p38 (SAPK) is crucial and essential to initiate myogenesis. It is suggested that the auto- and/or paracrine mechanism decisive for myogenesis leads to the activation of p38 MAPK with subsequent genomic effect. NF-κB induction points to the cytokines that improve cell viability, whereas elevated Bcl-2 protein prevents intrinsic apoptosis. Pawlikowska et al. (Reference Pawlikowska, Gajkowska, Hocquette and Orzechowski2006) showed that mitochondria are essential for insulin-mediated myogenesis. The mitofusin-2 protein derived from the hyperplasia suppressor gene (hsg) mediates this role of mitochondria by interacting with the RAS protein in C2C12 cells (Pawlikowska et al., Reference Pawlikowska, Gajakowska and Orzechowski2007a). Metabolic programming of muscles also proved to be sensitive to cachexia (Pawlikowska et al., Reference Pawlikowska, ̡okociejewska, Hocquette and Orzechowski2007b), further supporting the role of energy metabolism in myogenesis. Furthermore, IGF-I increased proliferation of C2C12 cells and simultaneous addition of IGF binding protein (IGFBP)-2 and -3 inhibited this effect probably due to IGF-I sequestering (Oksbjerg et al., Reference Oksbjerg, Nissen and Vestergaard2006).
Theil et al. (Reference Theil, Sørensen, Nissen and Oksbjerg2006) showed that IGF-I, IGF-II and type I IGF-receptor were upregulated in porcine satellite cells during differentiation. IGFBP-5 expression was greatly elevated as cells were grown to near confluence supporting the role of IGFBP-5 in inhibiting the mitogenic effects of IGF-I but stimulating its myogenic effect. Myostatin was also upregulated during myogenesis supporting its inhibitory role in controlling proliferation and differentiation. Exogenous insulin stimulated cell differentiation by upregulating IGFBP-5 and downregulating myostatin, whereas myogenin expression was increased. In a study with large and small littermate pigs in utero, Karunaratne et al. (Reference Karunaratne, Bayol, Ashton and Stickland2007) found elevated levels of IGFBP-5 in the small littermate and suggested this protein to be involved in the shift from muscle to connective tissue in developing muscles. Comparing proliferating with differentiating porcine satellite cell cultures under serum-free conditions Kalbe et al. (Reference Kalbe, Mau and Rehfeldt2008) observed a slight decrease in IGF-I and increases in IGF-II and epidermal growth factor (EGF) mRNA expression in differentiating cultures, whereas IGF receptor-1 expression tended to be upregulated. Fahey et al. (Reference Fahey, Brameld, Parr and Buttery2005a) hypothesised that the period of proliferation immediately before fibre formation would be crucial. In foetal sheep they noted a peak for myogenin and IGF-II mRNA expression on day 85 of gestation, which indicated that differentiation and formation of the majority of muscle fibres in the leg were occurring at this time.
In cattle, proteomic analysis of muscle at five key gestational stages revealed that the abundance of 245 proteins changed with myogenesis. Mainly, three categories of proteins were displayed: contractile proteins (many isoforms of myosin heavy chain (MyHC) and troponin T (TnT)), cytoskeletal proteins (desmin, actin, HSP27) and proteins involved in cell cycle regulation and/or in the control of the proliferation/apoptosis balance. This last group of about 25 proteins, including PAK2, PARK7, SNX6, desmin, galectin 1 and others, could lead to the identification of a potential marker of the TNF (Chaze et al., Reference Chaze, Meunier, Chambon, Jurie and Picard2008), which is one of the most important traits that control muscle mass. In cattle, TNF is fixed from 180 dpc onwards. This is supported by proteomic data indicating a reduction in cell proliferation (Chaze et al., Reference Chaze, Meunier, Chambon, Jurie and Picard2008). Thus, stathmin, which plays an important role in cell cycle regulation, showed decreased expression from 180 dpc onwards. Septin proteins (septin 2 and 11 isoforms) involved in cytoskeletal organisation, scaffolding and division plane had a stable expression up to 180 dpc, confirming intense cell division until this stage, and declined thereafter. In parallel, annexin A1, which has an anti-proliferative function via the activation of ERK pathways, showed increased abundance from 180 dcp onwards. Increased maturation is reflected by significant changes in the profiles of protein isoforms belonging to metabolic and contractile pathways. Thus, proteins of the aldehyde dehydrogenase family involved in the metabolism of amino acids, carbohydrates and lipids, dihydrolipoyl dehydrogenase, involved in pyruvate metabolism, as well as enolase, lactate dehydrogenase (LDH) and isocitrate dehydrogenase (ICDH), all enzymes involved in carbohydrate metabolism, showed increased abundance from 180 dpc onwards. Similar results were found for proteins of the contractile apparatus. Abundance of different isoforms of myosin light chains 1 and 2, α-cardiac actin, fast TnT (fTnT), myosin-binding protein and adult forms of MyHC increased during the later stages of foetal life (Chaze et al., Reference Chaze, Meunier, Chambon, Jurie and Picard2009). Proteomic analysis of cattle with high muscle growth potential (double-muscled with mutation in myostatin gene or selected on this criteria) revealed some proteins related to muscle mass (Picard et al., Reference Picard, Bouley, Cassar-Malek, Bernard, Renand and Hocquette2005). For example, the ratio of fTnT from exon 16 to fTnT from exon 17 was found to be a good marker of muscle mass development.
In chicken and turkey embryos, neuromuscular stimulation influenced the IGF axis gene expression in a muscle fibre type-dependent manner. Fibre number and oxidative capacity were also increased in the Musculus semitendinosus (ST) of the stimulated chicks before hatching (Heywood et al., Reference Heywood, McEntee and Stickland2005). Neuromuscular stimulation caused muscle phenotype-dependent changes in the expression of the IGFs and their binding proteins in developing slow and fast muscle of chicken embryos (McEntee et al., Reference McEntee, Simbi, Bayol, Macharia and Stickland2006). In addition, in chicken Musculus pectoralis the mRNA expression of the PTEN gene, an inhibitor of cell signalling, increased with age, whereas the protein levels of PTEN and its activity were considerably decreased between E18 and d43 stages, suggesting its involvement in myogenesis (Vaudin et al., Reference Vaudin, Dupont, Duchene, Audouin, Crochet, Berri and Tesseraud2006).
Changes in MyHC isoforms and enzymes representative of mitochondrial activity or glucose metabolism have been studied in cattle. These studies indicated a disappearance of foetal myosin and a gradual increase in oxidative enzymes from day 180 of gestational age, whereas enzymes of glucose metabolism progressively increased throughout gestation (Cassar-Malek et al., Reference Cassar-Malek, Picard, Kahl and Hocquette2007). Picard and Cassar-Malek (Reference Picard and Cassar-Malek2009) have demonstrated the expression of the IIb MyHC isoform in the ST and Musculus longissimus thoracis of a French beef breed, Blonde d’Aquitaine, for the first time. MyHC IIb was expressed at the mRNA level in the two muscles from the eleven young bulls studied (Picard and Cassar-Malek, Reference Picard and Cassar-Malek2009). Furthermore, the ST muscle exhibited higher proportions of MyHC I (slow oxidative isoform) and MyHC IIa (fast oxido-glycolytic isoform), a lower proportion of MyHC IIx (fast glycolytic isoform), higher cytochrome-c oxidase and β-hydroxyacyl-CoA dehydrogenase activities and a lower ratio of LDH/ICDH in cloned compared to control heifers at 8 months of age (Jurie et al., Reference Jurie, Picard, Heyman, Cassar-Malek, Chavatte-Palmer, Richard and Hocquette2009). Thus, young cloned heifers had slower muscle types associated with a more oxidative muscular metabolism than control heifers. From 12 months of age, significant differences were no longer observed between cloned and control heifers.
Rescan and co-workers used whole mount in situ hybridisation of rainbow trout with a large repertoire of somite-specific transcripts to describe the gene activations that underlie the somite patterning in teleost fish (Chauvigné et al., Reference Chauvigné, Cauty, Rallière and Rescan2005 and Reference Chauvigné, Rallière, Cauty and Rescan2006; Rescan, Reference Rescan2005 and Reference Rescan2008). They showed that the pax7-positive external cell layer that provides myogenic cell for the growth of the embryonic myotome also exhibits features of dermogenic differentiation as indicated by the expression of Dermo-1 (Dumont et al., Reference Dumont, Rallière and Rescan2008) and collagen (Rescan et al., Reference Rescan, Rallière, Chauvigné and Cauty2005). These expression patterns support the view that the external cell layer in fish is homologous to the amniote dermomyotome as suggested by Devoto et al. (Reference Devoto, Stoiber, Hammond, Steinbacher, Haslett, Barresi, Patterson, Adiarte and Hughes2006). It has also been shown that the NLRR-1 gene, which encodes a cell adhesion molecule regulating morphogenesis, is specifically expressed in migrating slow muscle cells (Dumont et al., Reference Dumont, Rallière, Tabet and Rescan2007), whereas the genes for the helix–loop–helix proteins, Id6a, Id6b, Id1 and Id2 – which are specific myogenesis inhibitors – are specifically expressed in the ventral and dorsal domains of the developing fish somites (Rallière et al., Reference Rallière, Chauvigné and Rescan2004).
Muscle development was also studied in teleost fish using the TnT and myosin light chain (MLC) genes as markers. Different isoforms mark different developmental stages and probably different developmental mechanisms as well as different muscle fibre types (Campinho et al., Reference Campinho, Power and Sweeney2005a, Reference Campinho, Power and Sweeney2005b and Reference Campinho, Silva, Nowell, Llewellyn, Sweeney and Power2007; Moutou et al., Reference Moutou, Silva, Mamuris and Power2005; Silva et al., Reference Silva, Campinho and Power2007a). In addition, these genes can be used as markers for the study of prenatal events with an effect on postnatal growth performance, for example, the axial musculature of the developing larvae of the Atlantic halibut is the largest and most rapidly growing tissue and during the transition from larval to adult muscle fibre types significant changes in fibre morphology and gene transcription occur. The change in myotome height correlates well with different larval halibut stages.
The mRNA expression of fTnT genes in the flatfish Atlantic halibut, which undergoes TH-driven metamorphosis leading to a dramatic change from a symmetrical larva to an asymmetrical juvenile and the roundfish sea bream, was studied throughout development and in adult tissue. Three alternative spliced forms of fTnT were identified. In both species, the isoforms appear to be stage-specific and correspond in sea bream to embryonic- (efTnTsb), larval- (LfTnTsb) and adult (afTnTsb)-specific isoforms, whereas in halibut they correspond to embryonic/larval (efTnThh) and juvenile/adult (fTnThh-1 and -2) isoforms. In the sea bream, each isoform is characteristic of the principal developmental stages and the embryonic acidic isoform (efTnTsb) is downregulated immediately after hatching and is replaced by a larval isoform, LfTnTsb. In contrast, in pre-metamorphic halibut larvae all three fTnThh isoforms are present although the most acidic form, efTnThh, is most abundant up until metamorphosis, after which it is downregulated to almost undetectable levels. At metamorphosis and in subsequent stages, fTnThh-2 is upregulated approximately threefold and becomes the most abundant isoform in halibut muscle. In contrast, fTnThh-1 expression is constant throughout development. In halibut, a novel teleost-specific fTnT-like cDNA (AfTnT) expressed exclusively in slow muscle was identified, which had a constant expression throughout ontogeny. The results from this comparative analysis of fTnTs in two teleosts suggests that molecular processing is common, but that regulation of isoform expression during muscle ontogeny may be species-specific and adapted to their specific ecologies (Campinho et al., Reference Campinho, Silva, Nowell, Llewellyn, Sweeney and Power2007; Silva et al., Reference Silva, Campinho and Power2007a).
In the sea bream, two slow skeletal muscle TnT genes (sTnT1sb and sTnT2sb) and a single form, sTnT2, were isolated and characterised. A third, intronless, TnT gene (iTnTsb), which is an apparent orthologue of a previously described zebrafish TnT, was also isolated in sea bream. The expression of sTnT was restricted to red muscle and sTnT1sb was expressed at low abundance and sTnT2sb at high abundance. In contrast, iTnTsb is predominantly expressed in adult fast muscle. The teleost sTnT2 and iTnT each constitute new, apparently teleost-specific, TnT groups. It remains to be established if the encoded proteins have different structural and mechanistic roles in fish muscle (Campinho et al., Reference Campinho, Power and Sweeney2005a, Reference Campinho, Power and Sweeney2005b and Reference Campinho, Silva, Nowell, Llewellyn, Sweeney and Power2007). Treatment of halibut larvae with THs, important post-embryonic morphogens, modified the expression of halibut fTnT isoforms, which changed from predominantly basic to acidic isoforms and led to the precocious downregulation of efTnThh (Campinho et al., Reference Campinho, Silva, Nowell, Llewellyn, Sweeney and Power2007; Silva et al., Reference Silva, Campinho and Power2007a). In contrast, the expression of red muscle-specific genes, AfTnT and sTnT2, in halibut did not change during natural metamorphosis or after T4 treatment (Campinho et al., Reference Campinho, Silva, Nowell, Llewellyn, Sweeney and Power2007). Treatment of sea bream larvae with THs revealed that although slow TnT1 is TH-responsive, fTnT, sTnT2 and iTnT expression are unaffected (Campinho et al., Reference Campinho, Power and Sweeney2005a and Reference Campinho, Power and Sweeney2005b).
At the early stages of sea bream development MLC1 and MLC2 were expressed exclusively in white muscle and no expression was observed in the superficial red muscle layer. On day 4 the expression of both transcripts was strong throughout the epaxial and hypaxial musculature. From day 10 onwards, two distinct germinal zones appeared in the dorsal and ventral side of the larvae, characterised by small diameter muscle fibres, whereas fibre diameter gradually increased from the lateral germinal zones towards the horizontal myoseptum. Fibre diameter of the deep white muscle layers next to the notochord indicated high hypertrophic activity. At the same time, MLCs’ expression became restricted to the periphery of the maturing muscle fibres and it was predominant at the germinal zones. This pattern persisted up to day 51, when the germinal zones disappeared and expression of MLCs was observed only in cells situated between the mature white fibres, most probably satellite cells (Moutou et al., Reference Moutou, Silva, Mamuris and Power2005). In gilthead sea bream, MLC2 (isoform A) consists of three mRNAs differing in length of the 3′ UTRs. Transcripts of isoform A were detected both in white and red muscle. A 1.56 kb MLC2 transcript was detectable from the beginning of somitogenesis, whereas a 0.89 kb MLC2 transcript was present after 27 h post-fertilisation. MLC2 isoform B differs from the isoform A by 10 amino acids. Isoform B was detected in all tissues examined (red, white, smooth and cardiac muscle, kidney, liver, spleen, brain, gills, epidermis; Sarropoulou et al., Reference Sarropoulou, Power, Mamuris and Moutou2006).
In the developing Atlantic halibut, first-feeding larvae MLC1, MLC2A and MLC3 transcripts were confined to the muscle fibres of the germinal zones located at the dorsal and ventral apices of the myotome and were composed of small-diameter round fibres. In pre-metamorphic larvae, transcripts were highly expressed throughout the epaxial and hypaxial musculature and expression levels reached a maximum in larvae starting metamorphosis coinciding with an increasing concentration of THs. However, at the metamorphic climax, MLC1, MLC2A and MLC3 expression was confined to fibres adjacent to the myosepts and to small cells scattered in the musculature, possibly satellite cells. MLC2A was also expressed in the red muscle fibres; no transition between larval and adult MLC isoforms was detected (Silva et al., Reference Silva, Baptista and Power2006). It will be of interest in the future to use the markers developed in sea bream and halibut to establish how changes in larval rearing conditions impact on subsequent muscle growth, fibre number and performance.
Environmental impact on skeletal muscle development and consequences for animal growth, carcass and meat quality
Prenatal and early postnatal nutrition
There is evidence for prenatal effects of nutrition on postnatal development of skeletal muscle in a variety of species, but mainly in young offspring of sheep, pigs, cattle and laboratory animals (Greenwood and Cafe, Reference Greenwood and Cafe2007; Brameld and Daniel, Reference Brameld and Daniel2008; Funston et al., Reference Funston, Larson and Vonnahme2010). To investigate the effects of in utero nutrition in mammals, two experimental models have been considered in this respect: first, the natural variation in birth weight of litter- or twin-bearing animals, which is most probably a result of different levels of nutrition in utero, and second, changes in maternal nutrition at different stages of pregnancy.
Comparison of high- and low-foetal weight or birth weight
Most of the studies concerning these issues have been done in pigs as there is a large natural variation in birth weight. During COST 925 the studies conducted by Bee (Reference Bee2004), Nissen et al. (Reference Nissen, Jorgensen and Oksbjerg2004), Karunaratne et al. (Reference Karunaratne, Ashton and Stickland2005), Gondret et al. (Reference Gondret, Lefaucheur, Juin, Louveau and Lebret2006), Lawlor et al. (Reference Lawlor, Lynch, O’Connell, McNamara, Reid and Stickland2007), Ruusunen et al. (Reference Ruusunen, Puolanne and Partanen2007a), Bérard et al. (Reference Bérard, Kreuzer and Bee2008), Rehfeldt and Kuhn (Reference Rehfeldt and Kuhn2006) and Rehfeldt et al. (Reference Rehfeldt, Tuchscherer, Hartung and Kuhn2008b) contributed substantially to our understanding of skeletal muscle development and long-term consequences for growth, body composition and meat quality in pigs. In summary, muscles of pig foetuses small for their gestational age or piglets light at birth (LW) have a lower number of fibres and lower total DNA content than middle (MW) or heavy weight (HW) offspring. During postnatal development, the smallest littermates deposit more fat and collagen at the expense of muscle tissue than the largest littermates. Intrauterine growth retardation (IUGR) together with impaired muscle development has long-term consequences for postnatal growth and later carcass and meat quality. Piglets small at birth grow slower and remain lighter at a certain age of slaughter or they need a longer time to reach a certain market weight compared with their heavier littermates. Importantly, these pigs develop lower carcass quality in terms of higher fat deposition and lower lean accretion compared with piglets of medium or heavy birth weights. This has been found to be more pronounced in female than in male pigs. In addition, lower meat quality in terms of tenderness scores, pH at 45 min post mortem, and water-holding capacity (drip loss) has been observed at slaughter in pigs of low birth weight, with optimum meat quality seen in medium birth weight pigs. As there were only marginal differences in fibre-type frequencies, this seems to be mainly related to the extreme fibre hypertrophy in LW pigs associated with greater MyHC expression, which occurs when the number of fibres is low. Extreme fibre hypertrophy also causes a higher incidence of giant fibres in post-mortem muscle, which is a clear sign of poor pork quality (Schubert-Schoppmeyer et al., Reference Schubert-Schoppmeyer, Fiedler, Nurnberg, Jonas, Ender, Maak and Rehfeldt2008). However, consistent with the overall higher fatness, the intramuscular fat was higher in LW than in HW pigs, which may even be of advantage in terms of eating quality.
There are only a few studies comparing sheep differing in birth weight by natural variation. Within our COST-related research Ensoy et al. (Reference Ensoy, Aksoy, Sirin, Sen, Ulutas and Kuran2008) were able to show that lambs of low birth weight (2.6 kg) deposited more fat during postnatal growth than lambs of high birth weight (4.1 kg), which is consistent with results previously obtained by Greenwood et al. (Reference Greenwood, Hunt, Hermanson and Bell1998). Previously, Greenwood et al. (Reference Greenwood, Slepetis, Hermanson and Bell1999 and Reference Greenwood, Hunt, Hermanson and Bell2000) did not observe differences in myofibre number, but decreased muscle weights, muscular DNA, RNA and protein contents and cell cycle activity at low foetal or birth weight. Similarly McCoard et al. (Reference McCoard, McNabb, Peterson, McCutcheon and Harris2000) found no differences in myofibre number comparing single and twin lamb foetuses. Moloney and Drennan (Reference Moloney and Drennan2006) studied the effects of birth weight on growth and carcass quality in Charolais cattle and found heavier carcasses that tended to be leaner from high birth weight (54.1 kg) compared with low birth weight (41.4 kg) calves, whereas meat quality characteristics remained unaffected. Papstein et al. (Reference Papstein, Losand, Fiedler, Hartung and Ender1999) have previously shown that the total number of fibres in M. longissimus was lower in twin compared with single-born beef cattle that were heavier at birth. Greenwood and Cafe (Reference Greenwood and Cafe2007) reported smaller weight and cross-sectional area of longissimus muscle in response to retarded intrauterine growth, but apparent fibre number at 30 months of age was not different. Similarly, no differences in beef quality were apparent. Taken together, these sheep and cattle studies suggest that restricted foetal growth resulting in low birth weight has similar consequences for postnatal growth rate and carcass composition as described in the pig and that this is related to myofibre number and/or muscular DNA content, whereas the effects on meat quality seem to be marginal. Species specificities in the nutritional foetal programming have been observed between ovine and bovine, due to differences in the timing of placental and foetal growth (Greenwood and Cafe, Reference Greenwood and Cafe2007)
In conclusion, impaired myogenesis resulting in low skeletal myofibre and myonuclear numbers is considered one of the main reasons for the negative long-term consequences of IUGR on muscle growth.
Consequences of maternal nutrition
Maternal diet controls growth directly by providing glucose, amino acids and other essential nutrients and metabolites for the conceptus (Robinson et al., Reference Robinson, Sinclair and McEvoy1999). In mammalian species these are transferred by passive and active transport through the placenta, which exhibits a central role for growth. Research during the COST action provided new insights into the effects and mechanisms of maternal nutrition on growth, mainly for pigs and sheep, but also for rodent models such as mice and rats. Both over- and undernutrition of energy and/or specific dietary compounds have been investigated. As recent results for pigs and sheep have been comprehensively reviewed by Brameld and Daniel (Reference Brameld and Daniel2008), only those results obtained during COST 925 will be specifically reviewed here.
High- and low-energy feeding
Researchers have hypothesised previously that an increase in a sow’s feed level may increase the growth potential of the pig (Dwyer et al., Reference Dwyer, Stickland and Fletcher1994). On the basis of this hypothesis several experiments have been conducted and analysed during the past few years. Thus, Bee (Reference Bee2004) fed multiparous sows low- (ca. 60%) and high-energy (ca. 140%) diets compared with a standard diet (100%) during the first 50 days of gestation. Progeny of the high-energy sows grew slower and had lower gain-to-feed ratios and higher percentages of adipose tissue (AT) than pigs born to control and/or low-energy sows. Similarly, Nissen et al. (Reference Nissen, Danielsen, Jorgensen and Oksbjerg2003) found a significant negative effect of ad libitum feeding of pregnant sows from days 25 to 50 on the average daily weight gain and muscle deposition rate. The effect on muscle deposition rate was most pronounced in the smaller littermates. Lawlor and co-workers (Lawlor and Lynch, Reference Lawlor and Lynch2005; Lawlor et al., Reference Lawlor, Lynch, O’Connell, McNamara, Reid and Stickland2007; McNamara et al., Reference McNamara, Stickland, Lynch, O’Connell and Lawlor2008) studied the effects of maternal energy intake of multiparous sows feeding at 200% of requirements between days 25 and 50 or days 50 and 80 and 150% from days 80 to 112 of gestation. Growth performance, such as daily weight gain, feed conversion efficiency and pig market weight, was not influenced, but carcass back fat thickness was reduced after feeding with 200% of requirements. Intra-litter variation and the overall sow performance were also unaffected. Increasing energy allowance between days 50 and 80 of gestation even increased the number of stillborn piglets per litter. In addition, Heyer et al. (Reference Heyer, Andersson, Lindberg and Lundstrom2004) found only marginal effects of increased maternal feed allowance from days 25 to 85 (+35%, +70%, +100%) on the growth, carcass quality and technological meat quality of the offspring. Birth weight and intra-litter variation were not influenced, whereas growth rate was negatively affected in the progeny of second parity sows, but these had higher numbers of weaned piglets. Cerisuelo et al. (Reference Cerisuelo, Sala, Carrion, Coma, Tibble, Gasa and Baucells2006, Reference Cerisuelo, Sala, Gasa, Chapinal, Carrion, Coma and Baucells2008 and Reference Cerisuelo, Baucells, Gasa, Coma, Carrion, Chapinal and Sala2009) studied the effects of extra feeding during mid-gestation in the pig. The controls received 3 kg/day (12 MJ ME/kg; 100%) throughout gestation and the experimental group received 150% (primiparous) or 175% (multiparous) from days 45 to 85 of gestation. This additional feed allowance had no beneficial effects on postnatal growth performance of the progeny. Total fibre number and numbers of primary and secondary myofibres were lower in M. longissimus after maternal supplementation, which was contrary to the expected higher (Dwyer et al., Reference Dwyer, Stickland and Fletcher1994; Gatford et al., Reference Gatford, Ekert, Blackmore, De Blasio, Boyce, Owens, Campbell and Owens2003) or unchanged (Nissen et al., Reference Nissen, Danielsen, Jorgensen and Oksbjerg2003) numbers. This was mainly due to lower numbers of IIB fibres and was associated with improved meat quality in terms of lightness and pH at 24 h post mortem, whereas carcass quality in terms of lean meat percentage or back fat thickness remained unchanged.
With regard to the underlying mechanisms, Nissen et al. (Reference Nissen, Sorensen, Vestergaard and Oksbjerg2005) showed that ad libitum compared with restrictive feeding of pregnant sows in early to mid-gestation did increase maternal IGF-I, but not glucose concentration, and did not increase either foetal glucose or IGF-I levels and had no beneficial effects on myogenesis. In agreement, there was no effect of maternal nutrition on the ability of foetal serum to stimulate proliferation and differentiation of primary myoblasts in vitro. When pregnant sows were administered GH during early gestation, which stimulated myofibre formation mainly in the small littermates, more glucose became directly available to the foetus (Rehfeldt et al., Reference Rehfeldt, Nissen, Kuhn, Vestergaard, Ender and Oksbjerg2004b). The increase in maternal IGF-I in response to maternal overfeeding is obviously not high enough to increase placental efficiency and thereby to act positively on myogenesis. In summary, additional energy supply to pregnant modern-day sows seems to have no beneficial effects on the growth of the progeny with myogenesis not or only slightly affected.
The results on low-energy feeding of pregnant sows (Bee, Reference Bee2004) largely confirmed previous studies showing that a mild maternal feed restriction has marginal impact on the growth of the offspring, whereas severe maternal feed restriction is known to impair myogenesis and subsequent postnatal growth (Buitrago et al., Reference Buitrago, Walker, Snyder and Pond1974).
Comprehensive studies on the influence of maternal nutrition have been conducted by Brameld et al. (Reference Brameld, Fahey, Langley-Evans and Buttery2003 and 2008) in sheep. The hypothesis being tested was that the critical window for the effects of maternal nutrition on muscle fibre numbers related to the period of proliferation immediately before major fibre formation. The first study (Fahey et al., Reference Fahey, Brameld, Parr and Buttery2005a), therefore, set out to determine the timing for major fibre formation in foetal sheep in order to identify the periods of proliferation and differentiation prior to major fibre formation. A peak for myogenin and IGF-II mRNA expression was observed on day 85 of gestation, which, together with histochemical analyses, indicated that the differentiation and formation of the majority of muscle fibres in the leg were occurring at this time. The authors then tested their hypothesis by comparing offspring from adequately fed, control ewes with those from ewes that were nutrient restricted (50% of intake of controls) before (days 30 to 70), during (days 55 to 95) or after (days 85 to 115) the peak in fibre formation (Fahey et al., Reference Fahey, Brameld, Parr and Buttery2005b). In agreement with the hypothesis, maternal undernutrition had no effect during (days 55 to 95) or after (days 85 to 115) major fibre formation in 14-day-old lambs (Fahey et al., Reference Fahey, Brameld, Parr and Buttery2005b). The critical period was before the peak in fibre formation (days 30 to 70), with maternal undernutrition resulting in reduced secondary-to-primary fibre ratio. Later studies investigated the long-term effects of maternal undernutrition targeted at this critical window. Despite the fact that previous studies targeting the period immediately before major fibre formation had shown effects on muscle fibre development in neonatal lambs (Fahey et al., Reference Fahey, Brameld, Parr and Buttery2005b), when sheep were grown to conventional slaughter weights (40 to 45 kg or 17 weeks of age) or beyond (24 weeks; Daniel et al., Reference Daniel, Brameld, Craigon, Scollan and Buttery2007), those differences were lost and even a slight increase in proportions of fast glycolytic (IIx/b) fibres observed in older animals (24 weeks). However, maternal undernutrition was associated with increased measures of adiposity in older sheep, with no effect on food intake, suggesting a possible small long-term reduction in metabolic rate. In addition to the presented sheep studies, Harrison and co-workers found that severely reduced nutrient availability during late gestation affected growth of the lambs negatively and had prolonged negative effects on lactation performance (Tygesen et al., Reference Tygesen, Harrison and Therkildsen2007 and Reference Tygesen, Nielsen, Nørgaard, Ranvig, Harrison and Tauson2008) and changed muscle metabolism in young offspring measured by evoked surface electromyography (SEMG; Tygesen and Harrison, Reference Tygesen and Harrison2005).
In the rat, maternal undernutrition to 50% and 40% of ad libitum had opposite impacts on the offspring’s postnatal growth rates with the most severe reduction (40%) leading to slower growth and fewer muscle nuclei (Bayol et al., Reference Bayol, Jones, Goldspink and Stickland2004). More recent studies (Bayol et al., Reference Bayol, Simbi and Stickland2005 and Reference Bayol, Farrington and Stickland2007 and Reference Bayol, Simbi, Bertrand and Stickland2008) have used a ‘junk food’ model in which pregnant rats were given free access to processed human foods rich in fat, sugar and salt. The offspring of these mothers grew heavier, produced more fat and developed a greater appetite and taste for junk food themselves when compared to offspring of control mothers fed a balanced diet. The offspring from junk-food-fed mothers also had atrophied muscles with more intramuscular fat at weaning.
High- and low-protein feeding
Mallinson et al. (Reference Mallinson, Sculley, Craigon, Plant, Langley-Evans and Brameld2007) investigated the effects of feeding a maternal low-protein diet at different stages of pregnancy in the rat on the numbers, proportions and diameters of muscle fibres in the resulting pups at 4 weeks of age. The critical window for rats was mid-to-late pregnancy, when most of the muscle fibres start to form. Thus, maternal low-protein (LP) diet mainly during mid-pregnancy reduced muscle fibre number and density, but there were muscle-specific differences in the fibre types. Metges and co-workers (Metges, Reference Metges2005; Petzke et al., Reference Petzke, Daenzer-Wiedmer, Günther, Lang, Kucia, Klaus and Metges2008) have shown in a rat model that both maternal LP and high-protein (HP) intake throughout pregnancy resulted in low birth weight of the pups and increased body fatness in adolescent offspring of HP dams. The results suggested that oxidative losses of amino acids induced by HP feeding may have similar consequences for placental and foetal protein synthesis as the lack of maternal dietary amino acids, both resulted in foetal growth retardation.
In a mouse model, females of three different strains (selected for long distance running (LDR), for high body weight (HBW) or unselected controls (CON)) were exposed to three prenatal–postnatal dietary combinations: HP-C (high-protein control diet), C-HP and C-C (Langhammer et al., Reference Langhammer, Derno, Dietrich, Renne, Nürnberg, Hennig and Metges2006; Kucia et al., Reference Kucia, Langhammer, Dietrich, Derno, Renne, Nürnberg, Hennig and Metges2007; Metges et al., Reference Metges, Dietrich, Kucia, Langhammer, Nürnberg and Rehfeldt2008). Birth weight was reduced in response to HP diet throughout pregnancy in the CON and LDR lines, but not in the HBW line. Effects on muscle cellularity are currently being investigated. The offspring of all lines grew slower in response to the HP lactation diet. Overall, the results showed that programming response due to maternal HP diet in mice is genotype-dependent.
While detrimental effects of postnatal LP diets on growth and carcass quality in pigs are well known and have been confirmed recently (Ruusunen et al., Reference Ruusunen, Partanen, Pösö and Puolanne2007b), little is known about the effects of different protein levels in the maternal sow diet. Preliminary results on maternal HP (250% above) and LP (50% below) feeding throughout gestation to sows also reveal that birth weight was reduced in response to both diets and is thus detrimental to foetal growth, and were associated with changes in metabolism, body composition and muscular gene expression (Metges and Rehfeldt, Reference Metges and Rehfeldt2006; Gondret et al., Reference Gondret, Kalbe, Metges and Rehfeldt2008; Rehfeldt et al., Reference Rehfeldt, Kalbe, Block, Nuernberg, Stabenow, Loesel and Metges2008a).
Specific nutrients
Due to the effects of maternal undernutrition on muscle cell proliferation/differentiation observed in the various in vivo studies, Brameld and co-workers have also investigated direct effects of nutrients on muscle differentiation in vitro (Hurley et al., Reference Hurley, Flux, Salter and Brameld2006). They found that a number of unsaturated fatty acids (both poly- and mono-unsaturated) altered differentiation of L6 muscle cells, with no effect of palmitic acid, the only saturated fatty acid studied. This may be a mechanism for the effects of maternal diet on muscle development.
Soya-based formulas are used as regular food supplements in livestock production and are consumed by growing numbers of infants. The effects of constituent isoflavones such as daidzein and/or genistein have been examined when supplemented to sows during gestation as well as in a porcine muscle satellite cell culture model. Supplemental daidzein (leading to ca. 1 μmol/l in serum, similar to a typical diet containing soyabean meal) in the maternal diet during late gestation (from day 85 to term) marginally affected carcass quality, meat quality and skeletal muscle cellularity of the progeny (Rehfeldt et al., Reference Rehfeldt, Adamovic and Kuhn2007). The in vitro studies on the direct effects of isoflavones on muscle cells may be a model for the effects of maternal diet on muscle development. Results revealed that genistein and daidzein act as inhibitors of porcine myoblast proliferation at concentrations of >1 and >10 μM, respectively (Mau et al., Reference Mau, Kalbe, Viergutz, Nürnberg and Rehfeldt2008a), and that they interact with growth factor (IGF-I; EGF)-stimulated porcine muscle cell proliferation (Mau et al., Reference Mau, Kalbe, Wollenhaupt, Nürnberg and Rehfeldt2008b; Kalbe et al., Reference Kalbe, Mau and Rehfeldt2008). Reduced proliferation, in part, resulted from the inhibition of the expression of these growth factors and their receptors. Both genistein and daidzein also affected protein metabolism of porcine myotube cultures in a dose-dependent manner (Rehfeldt et al., Reference Rehfeldt, Kalbe, Nürnberg and Mau2009), reducing protein synthesis at high concentrations (100 μmol/l), but decreasing protein degradation rates at low concentrations, thereby potentially acting as stimulators of protein accretion.
Cross-talk between muscle and adipose cells
As described in previous sections, maternal undernutrition in mammals may reduce muscle hypertrophy and predispose offspring to lipid accumulation, which suggests either a prioritisation or competition in the nutrient partitioning. These adaptations could have postnatal consequences on the lean-to-fat ratio as the interplay between muscle and adipose growth constitutes an important factor in the regulation of muscle development well documented in Bonnet et al. (Reference Bonnet, Cassar-Malek, Chilliard and Picard2010).
The lean-to-fat ratio is the result of a dynamic balance between the number and size of muscle and adipose cells. There is striking evidence for developmental and functional links between muscle and AT – in bovine foetuses the successive waves of growth of muscle and AT suggest a priority for muscle growth. In double-muscled Belgian blue cattle, increased muscular development is concomitant with a decrease in the carcass and muscular AT. Thus, adipose and muscle cells seem to be linked by competition or prioritisation in their commitment, differentiation and/or uptake and metabolism of nutrients, all of which determine their number and size (Bonnet et al., Reference Bonnet, Cassar-Malek, Chilliard and Picard2010). In addition to the metabolic and endocrine regulations specific for each tissue, a cross-talk between adipose and muscle cells that leads to the reciprocal regulation of their differentiation has been deciphered by in vitro studies. In the COST action, models of primary co-culture and of conditioned medium have been used to study the direct interactions that occur during myogenesis and adipogenesis in bovine foetuses. One study revealed that myogenesis was impaired by proliferative preadipocyte-conditioned media as shown by a low fusion index, an indicator of myoblast fusion (Cassar-Malek et al., Reference Cassar-Malek, Bonnet, Chilliard and Picard2006). Conversely, myogenesis was enhanced by adipocyte-conditioned medium through an increase in the myotube area. These cross-talks could be mediated at least, in part, by adipokines such as leptin (Bonnet et al., Reference Bonnet, Cassar-Malek, Chilliard and Picard2010). Furthermore, the differentiation of foetal brown preadipocytes was enhanced by co-culture with primary bovine myoblasts. This may result both from a regulation by paracrine signals and from physical cellular interactions as already described for white adipocytes and endothelial cells. Among the paracrine signals that have been identified myostatin has been shown to impair the differentiation of bovine white preadipocytes by inhibiting the expression of PPARγ and C/EBPα. These data highlight the complexity of the cross-talk between muscle and adipose cells, as they involve both secreted factors and cell–cell interactions, and concern different cell types depending on the chronology of the growth of muscles and ATs.
Nutritional influences on muscle development in fish and poultry
The environment determines the rate of myogenesis, the composition of sub-cellular organelles, patterns of gene expression (Fernandes et al., Reference Fernandes, MacKenzie, Wright, Steele, Suzuki, Kinghorn and Johnston2006) and the number and size distribution of muscle fibres in fish (reviewed by Johnston, Reference Johnston2006). Knowledge about the influence of nutrition on muscle development and growth is still rudimentary in fish, but was explored during the COST action by Silva et al. (Reference Silva, Valente, Galante, Andrade, Monteiro and Rocha2009b) who analysed muscle cellularity in blackspot sea bream (Pagellus bogaraveo, Brunnich) juveniles fed on diets with different protein contents (20% to 60%). Fish fed HP diets (>40%) had higher final BWs and total muscle weights due to increased fibre area and fibre number, independent of muscle type and localisation. In Atlantic salmon, muscle cellularity and flesh quality were relatively insensitive to the dietary protein-to-energy ratio (Johnston et al., Reference Johnston, Manthri, Alderson, Campbell, Mitchell, Whyte, Dingwall, Nickell, Selkirk and Robertson2002; Björnsson et al., Reference Björnsson, Steinarsson and Árnason2007).
Several in vivo and in vitro studies have explored the impact of feeding regime on muscle growth. In rainbow trout fed different ration levels, Kiessling et al. (Reference Kiessling, Storebakken, Asgard and Kiessling1991) reported an increased importance of hypertrophy during periods of rapid growth. Starvation and re-feeding of brown trout (Salmo trutta) induced changes in muscle composition and muscle structure (smaller mean muscle fibre diameter in starved fish) that contributed to an increase in flesh firmness (Bugeon et al., Reference Bugeon, Lefevre and Fauconneau2004). Recovery growth in muscle in response to fasting and re-feeding was studied in trout using a cDNA microarray and revealed a complex response. The principal groups of upregulated transcripts were involved in protein catabolism, myofibre and muscle remodelling and regulation of transcription (Rescan et al., Reference Rescan, Montfort, Rallière, Le Cam, Esquerré and Hugot2007). Other studies of re-feeding in fish have shown enhanced muscle expression of kinases involved in the mTOR signalling pathway known to promote protein accretion in mammalian muscle (Seiliez et al., Reference Seiliez, Gabillard, Skiba-Cassy, Garcia-Serrana, Gutierrez, Kaushik, Panserat and Tesseraud2008).
The mechanism by which nutrition and feeding regime affect muscle growth in fish remains to be established but almost certainly involves modification of endocrine factors such as GH and IGF, which have a recognised role in regulating muscle mass, fibre size and nutrient partitioning (Reinecke et al., Reference Reinecke, Bjornsson, Dickhoff, McCormick, Navarro, Power and Gutiérrez2005; Codina et al., Reference Codina, Montserrat, Garcia de la Serrana, Rallière, Gabillard, Navarro, Rescan and Gutierrez2006). For example, in both rainbow trout and gilthead sea bream, fasting increased plasma GH levels and decreased tissue expression and plasma levels of IGFs and upregulated IGF-I binding (Gabillard et al., Reference Gabillard, Kamangar and Montserrat2006a; Montserrat et al., Reference Montserrat, Gabillard, Capilla, Navarro and Gutiérrez2007a and Reference Montserrat, Gomez-Requeni, Bellini, Capilla, Perez-Sanchez, Navarro and Gutierrez2007b). In trout, fasting followed by re-feeding causes a switch to fast growth and involves the local upregulation of several components of the IGF system (Gabillard et al., Reference Gabillard, Kamangar and Montserrat2006a). Moreover, IGF-I stimulates cell proliferation in both rainbow trout (Castillo et al., Reference Castillo, Codina, Martinez, Navarro and Gutierrez2004 and Reference Castillo, Ammendrup-Johnsen, Codina, Navarro and Gutierrez2006) and gilthead sea bream (Montserrat et al., Reference Montserrat, Sánchez-Gurmaches, García de la Serrana, Navarro and Gutiérrez2007c) muscle cells in vitro and IGF-II has both mitogenic and metabolic effects in trout muscle cell cultures (Codina et al., Reference Codina, de la Serrana, Sanchez-Gurmaches, Montserrat, Chistyakova, Navarro and Gutierrez2008).
In chicks, 2 days delayed feeding after hatching decreased satellite cell proliferation in Pectoralis major muscle and lowered the expression of neonatal isoform relative to embryonic isoform of MyHC (Berri et al., Reference Berri, Godet, Haj Hattab and Duclos2006). The deprivation of nutrients during this period also prevented the postnatal increase in the expression of IGF-I and of myostatin in skeletal muscle (Guernec et al., Reference Guernec, Chevalier and Duclos2004). Furthermore, it was shown that the effect of insulin (and probably IGF-I as well) on protein metabolism was exerted through stimulation of S6K1 activity. Tissue-specific regulation of the S6K1 gene has been shown (Duchene et al., Reference Duchene, Audouin, Berri, Dupont and Tesseraud2008a). In chicken muscle, the AKT and the ERK1/2 MAPK pathways appear instrumental in this regulation, despite the fact that IRS-1 and PI3K did not respond to insulin stimulation in the muscle tissue (Duchene et al., Reference Duchene, Audouin, Crochet, Duclos, Dupont and Tesseraud2008b).
Influence of physical factors
During the COST action some research teams investigated the influence of selected physical factors, including temperature, neuromuscular stimulation, and oxygen levels on muscle development in fish and/or poultry.
In fish, water temperature has a significant effect on the growth process of axial white muscle (hyperplasia/hypertrophy) during early stages that can affect the subsequent white muscle growth process of juveniles. Nonetheless, not all processes associated with myogenesis increase to the same extent as temperature rises, resulting in differences in muscle cellularity (reviewed by Stickland et al., Reference Stickland, White, Mescall, Crook and Thorpe1988; Johnston, Reference Johnston2006). As mosaic hyperplasia is the main contributor to overall muscle growth, the developmental stages at which this process is activated may be critical for subsequent muscle growth. If the proliferative capacity of myogenic cells is affected at such stages, this could affect the growth potential of larvae, since both the ultimate size and growth rate are related to the number of muscle fibres in young fish (Weatherley, Reference Weatherley1990). Atlantic salmon embryos grown at 5°C and 10°C until first feeding and at 5°C thereafter exhibited significant differences in muscle cellularity and by 21 weeks larvae maintained at 5°C were longer and heavier (Albokhadaim et al., Reference Albokhadaim, Hammond, Ashton, Simbi, Bayol, Farrington and Stickland2007). The authors suggested that the differences might be explained by the influence of temperature on appetite and activity. In rainbow trout juveniles, the temperature-induced increase in growth rates was associated with high plasma GH and IGF-1 levels (Gabillard et al., Reference Gabillard, Weil, Rescan, Navarro, Gutierrez and Le Bail2005). Moreover, the temperature-induced increase in muscle growth was associated with an increase in expression of GH receptor 1 (Gabillard et al., Reference Gabillard, Yao, Vandeputte, Gutierrez and Le Bail2006b) suggesting that this receptor has a key role in the response to temperature. In European sea bass, early thermal environments influenced white and red muscle hyperplasia in later stages suggesting the existence of early programming of axial white muscle hyperplasia (Alami-Durante et al., Reference Alami-Durante, Olive and Rouel2006a, Reference Alami-Durante, Rouel and Kentouri2006b and Reference Alami-Durante, Olive and Rouel2007). For example, white muscle hypertrophy was stimulated at 20°C and hyperplasia at lower temperatures in embryos, whereas both hypertrophy and hyperplasia were stimulated at 20°C in free-swimming larval sea bass (Alami-Durante et al., Reference Alami-Durante, Olive and Rouel2006a). In contrast, in blackspot sea bream P. bogaraveo (Brunnich) high temperatures (18°C v. 14°C) accelerated both embryonic (hatching) and pre-larval (mouth opening) development (Silva et al., Reference Silva, Rocha, Olmedo, Alvarez-Blàzquez, Galante and Valente2007b) but failed to modify cross-sectional fibre area or total number of white fibres at hatching. The non-uniform behaviour of different species of fish highlights the need for more studies to characterise this numerous but divergent class of vertebrates.
In rainbow trout, low oxygen saturation levels affect carcass quality with an increase in the proportion of red muscle in the muscle mass, although there is a minimal effect on flesh quality even at low oxygen levels (76% saturation). Stress before slaughter decreased the mean diameter of red muscle fibres, flesh colour, mechanical resistance and pH, but no major interaction between rearing oxygen level and slaughter stress was formally demonstrated in rainbow trout (Lefevre et al., Reference Lefevre, Bugeon, Auperin and Aubin2008).
In turkeys, it was demonstrated that small differences in incubation temperature for a few days early in development significantly influenced post-hatch muscle growth. A regime of 38.5°C caused an increased myonuclear and fibre number in ST, along with a delay in differentiation (Maltby et al., Reference Maltby, Somaiya, French and Stickland2004). In chickens, Hammond et al. (Reference Hammond, Simbi and Stickland2007) have shown that these small temperature manipulations alter muscle cellularity, possibly through an effect on in ovo movement activities (higher temperatures produced sustained increases in movement). Werner and Wicke (Reference Werner and Wicke2008) reported that increased temperature (38.5°C) between embryonic days 7 and 10 of incubation positively influenced the post-hatch muscle development in cocks, but not in hens. Heat conditioning during embryonic development in chickens has also been considered as a possible tool to favour the adaptation of birds to subsequent heat stress. Collin et al. (Reference Collin, Berri, Tesseraud, Requena Rodon, Skiba-Cassy, Crochet, Duclos, Rideau, Tona, Buyse, Bruggeman, Decuypere, Picard and Yahav2007) and Tona et al. (Reference Tona, Onagbesan, Bruggeman, Collin, Berri, Duclos, Tesseraud, Buyse, Decuypere and Yahav2008) showed that such treatment has no detrimental effects on muscle growth and meat quality. In summary, environmental temperature is a critical factor for growth and development of skeletal muscle in both fish and poultry. The mechanisms by which temperature exerts these effects remain to be identified.
Acknowledgements
We thank the COST organisation and all participants who contributed to the success of COST action 925.
