Skip to main content

Early control of growth for fish production with special reference to muscle development, gene expression and temperature


The main objective is to assess the effect of environmental temperature on muscle development and growth efficiency during early life. Emphasis will be drawn on muscle cellularity, differentiation, gene expression, and growth during the embryonic and larval stages in two main farmed species of fish: rainbow trout and sea bass.
Application of such temperature effects will be investigated through the consequences of alteration in muscle phenotypes by early temperature treatment on growth efficiency during post-larval growth in rainbow trout and in sea bass.
A complete description of the temperature effect during early development will be made to open the possibility of application for aquaculture.
Survival, growth and developmental rate of rainbow trout and sea bass embryos reared during early development at lower temperature than those usually practice have been followed. In rainbow trout differential mortality was observed at low temperature (4°C) and high temperature (12°C) tested. This seemed to be dependent on the strains and a shift towards higher temperature in the range of viable temperature for early development could be suspected in some rainbow trout strains. In sea bass preliminary experiments were realised a small scale in order to define the low temperature protocol (temperature, timing) to be applied in a large-scale experiment. The large-scale sea bass experiment is in progress and similar survival were observed at the three temperature tested. These experiments provide original data on temperature and early development in two farmed species.

A comprehensive analysis of some developmental stages was made in rainbow trout and sea bass as the timing of these stages was greatly affected by temperature. Description and timing of muscle development was also analysed in rainbow trout (see original results presented here). Serial sampling at specific developmental stages was made in rainbow trout and will be made in sea bass depending on temperature. In rainbow trout developmental rate was lowered at low temperature. Differences in body weight and length growth rates were also observed depending on temperature and consequently morphology of larvae was affected by temperature.

Muscle development will be followed by changes in cellularity (number and size of myotubes and fibres) and in expression of muscle specific genes (myogenic factors and contractile proteins). The muscle cellularity analysis is still in progress in rainbow trout. The lowering in developmental rate observed at low temperature for some strains required further analysis of the number and size of fibres to conclude on the positive effect of low temperature. In situ hybridisation methods were adapted to these early stages.

Specific rainbow trout cRNA probes from available cDNA of growth factors, myogenic factors and contractile proteins were tested successfully. In sea bass specific cDNA probes for growth factors, myogenic factors and contractile proteins were raised and started-to-be-tested. Thus new tools, only available up to now in zebra fish, have been developed in two farmed species: rainbow trout and sea bass for developmental biology studies.

Growth performances, muscle growth and muscle characteristics of fish previously exposed to different temperature regime during early development will be analysed from first feeding stage up to commercial stage (300 g) in rainbow trout. This experiment is in progress in two strains and in two different environmental conditions in rainbow trout.

Results and discussion

Table 1: Description of sampling stages (degree.d:degree.days).
(For Table 1 contact the Coordinator)

The different stages observed by direct microscopy were checked by SEM analysis (table 1). The rate of somitogenesis was 0.8 somites/ and it was similar at the two temperature studied. The shape of the somite changed from a pillow shape in the anterior part of the embryo at stage 16 up to a classical chevron shape observed all along the vertebral axis at eyed stage. During the embryonic period there was a 9-fold increase in the height of the somite (from 70 µm at stage 16 up to 600 µm at hatching), only a 3-fold increase in depth (from 30 to 90 µm) and few changes in width (lateral view from 30 to 60 µm).

On transverse fractures of embryos, thin elongated structures assimilated to myotubes were observed in the somite at stage 24. These structures seemed to be only present in the deep part of the somite, close to the notochord. At stage 26, cross-striations were observed in these structures. At hatching (stage 30), fibres composed of myofibrils (1 µm) were observed. The cross-striation was clearly distinguishable and sarcomere length was assessed to 1 µm.
Myosin expression was first observed at stage 20 using a fast MHC antibody S48E6 but not at early stages (16, 18). The expression was limited to the very deep part of the somite near the notochord at stage 20 and 22 (Table 2). Then at stages 24 and 26, the area of myosin expression expanded laterally with a gradient of intensity from the deep part of the somite to the periphery in agreement with the gradient in muscle differentiation described in zebra fish (Waterman 1969, Devoto et al. 1996). Myosin expression was observed in the whole somite at hatching (stage 30). The same results although more specific were observed with a fast MHC antibody S4 10H9 except that no expression was observed until stage 22.
No myosin expression was observed until stage 24 using a BA-D5 slow MHC antibody. At stage 24 (eyed stage) slow myosin expression was localised at the periphery of the somite in elongated mono-layer cells forming a U shape at the horizontal septum. The expression of slow myosin expanded dorsally and ventrally at stage 30 (hatching). Such differential expression of myosin isoforms has never been described in salmonids but it was suspected by early TEM analysis (Nag and Nursall 1972). This pattern is different to that observed in zebrafish where slow myosin expression start in the deep part of the somite (Devoto et al. 1996).
Table3: Fluorescent immuno labelling of fast and slow myosin in the somite during embryonic development (stages according to Vernier 1969).

(For Table 3 contact the Coordinator)

Desmin expression was observed very early in the deep part of the somite at stages 20, 22 and 24 with a similar pattern as that of the fast myosin antibody (S48E6) although intensity of labelling was more intense. No differences were observed in terms of myosin or desmin expression between low and high temperature at stages 24, 26 and 30.


Specific stages for muscle differentiation were identified. These results demonstrated also the role of notochord in the white muscle differentiation in rainbow trout. A sequential expression of fast and slow myosin with late expression of slo myosin in superficial cells of the somites with a different pattern as that of other models such as zebrafish.


Ballard, W. W. 1973. Normal embryonic stages for salmonids fishes, based on Salmo gairdneri (Richardson) and Salvelinus fontinalis (Mitchill). J. Exp. Zool. 184: 7-26.
Devoto, H. S., Melancon, E., Eisen, J. and Westerfield, M. 1996. Identification of separate slow and fast muscle precursor cells in vivo, prior to somite formation. Development. 122: 3371-3380.
Fauconneau, B., Paboeuf, G. 1998 Etude histoimmunologique de l'expression des chaines lourdes de myosine dans le muscle squelettique chez la truite arc en ciel. INRA Prod. Anim. (in press)
Henneguy, F. 1888. Embryogenese de la truite. J. Anat. Physiol. 24: 412-502; 526-617.
Nag A.C. Nursall, J.R. 1972 Histogenesis of white and red muscle fibres of trunk muscles of a fish Salmo gairdneri. Cytobios, 6: 227-246.
Pasteels, J. 1936. Etude de la gastrulation des vertebres meroblastiques. 1. Teleosteens. Arch. Biol. 47: 205-308.
Schiaffino, S., Gorza, L., Sartore, S., Saggin, L., Ausoni, S., Vianello, M., Gundersen, K. and Lomo, T. 1989. Three myosin heavy chain isoforms in type 2 skeletal muscle fibres. J. Muscle. Res. Cell. Motil. 10: 197-205.
Vernier, J. M. 1969. Table chronologique du developpement embryonnaire de la truite arc-en-ciel, Salmo gairdneri Rich. 1836. Ann. Embryol. Morphogenese. 4 (2): 495-520.
Waterman, R. E. 1969. Development of the lateral musculature in the teleost Brachydanio rerio: a fine-structural study. Am. J. Anat. 125: 457-494.
Embryonic muscle development was studied in rainbow trout (Oncorhynchus my kiss) at low and high temperature using scanning electron microscopy (SEM) and immuno-histology. Somites development was described starting at stage 16 and up to hatching stage. Somite growth in height was more important than in width and depth. Myotubes without any structure then myotubes with cross-striated contractile elements organised in myofibrils were successively observed in the somite. Immuno-histology analysis demonstrated the appearance of an embryonic fast myosin in the deep part of the somite at stage 20 (Vernier, 1969). The area of expression of myosin then expanded in the somite and covered the whole somite at hatching (stage 30). Slow myosin was only expressed at eyed stage (stage 24) in few superficial cells forming a U-shape at the horizontal septum and then expanded dorsally and ventrally. Specific stages for muscle development were thus identified in rainbow trout.


Early development of rainbow trout is well characterized anatomically and the mains stages have been fully described (Henneguy, 1988; Pasteels, 1936; Vernier, 1969; Ballard, 1973). Very little attention have been paid to muscle development apart from the number of somites used to identify the embryonic developmental stage. Analysis of the characteristics of cells within the somite in trout demonstrate an anteroposterior gradient in the development of skeletal muscle along the vertebral axis starting with the appearance of deep " white cells " and superficial "red cells" (Nag & Nursall, 1972) differentiating then in myotubes then in fast and slow fibres. Differentiation from early myotubes to muscle fibres in the deep part of the somite have been also characterized in herring embryos (Johnston & Viera 1995).

In the zebra fish (Danio rerio), used as a model for developmental studies new information about muscle development and myogenesis inside the somite have been obtained. Myogenesis initiates in the deep part of the somite near the notochord where two distinct populations of muscle precursors have recently been identified (Devoto et al., 1996). These precursor cells express different growth factors and myogenic factors as they became myoblasts. After a proliferating phase, myoblasts fusion leads to the formation of myotubes, differentiating later into muscle fibres.

The aim of this study was to describe morphological and functional aspects of muscle development during the embryonic stages of rainbow trout. Myogenesis was analysed through the expression of myosin inside the somite.


Fertilization were realized on a pool of ovules from 2 females and milt from 9 males. Eggs were incubated at two constant temperatures 12°C and 4°C and constant oxygen concentration (98 % saturation) up to end of yolk sac resorption.
Samples were collected every 50 degree.days at the two temperatures from fertilisation up to eyed stages then at the same developmental stages according to Vernier 1969. Eggs were dissected, embryos removed from the chorion. Direct observation was made by microscopy. Other embryos were treated for SEM or histoimmunology.
Scanning Electron Microscopy (SEM): Embryos were fixed in Karnovsky fixative (paraformaldehyde 4%, glutaraldehyde 5% sodium cacodylate 0.08 M) during 6 to 20 hours at 4°C then in osmium tetroxyde 0.1%. Samples were dehydrated by successive immersions in ethanol 30% to 100%. Samples were dried up to the critical point method and an or-palladium mixture was pulverized on the samples. They were observed using a scanning electron microscope (JEOL 8300, University of Rennes MEB centre) directly and after removal of the skin and fractures of the embryos.
Immunohistology : Embryos were fixed in ethanol 70% (in glycine buffer 0.05M pH = 2.0) then dehydrated (ethanol 95%, butanol) and paraffin embedded. Thin transverse section (5 µm) of the embryos were made and put on TESPA treated glass for immunology analysis (figure 1). Sections were desembedded, rehydratated (acetone, ethanol 100% to 70 %) and rinsed with distilled water and PBS 0.01M pH=7.4. Sections were then treated with BSA (0.2%) and saponine 0.2% in PBS. The section were treated with first antibody (20 µl/section) during 1 hour, rinsed with PBS, and with the second antibody coupled with FITC dye (20 µl/section anti mouse TEBU M 30001 anti rabbits TEBU L 42001) during 1 hour. The sections were rinsed with PBS, mounted in Mouwiol solution and observed with a light fluorescent microscope.


Institut National de la Recherche Agronomique
Campus De Beaulieu, Av. Du General Leclerc
35042 Rennes

Participants (4)

Main Port Heraklion
Royal Free Hospital School of Medicine
United Kingdom
Rowland Hill Street
NW3 2PF London
United Kingdom
Hawkshead Lane, North Mymms
AL9 7TA Welham Green,hatfieid
Universidade do Porto
Praça Gomez Teixeira
4050 Porto