Endocrinology and Metabolism

A proposed nomenclature consensus for the myostatin gene family

Buel D. Rodgers, Eric H. Roalson, Gregory M. Weber, Steven B. Roberts, Frederick W. Goetz

ever since its discovery in 1997 (11), myostatin and its negative effects on skeletal muscle mass have understandably captivated many biomedical, agricultural, and comparative biologists, since the gains in muscle mass associated with the myostatin null phenotype have never been reproduced by the administration of growth promoters regardless of species or mode of administration (9). The potential benefits of reproducing these effects in the clinic or in animal feed lots are obvious and cannot be overestimated. Relieving myostatin's restrictive effects on skeletal muscle growth and development could revolutionize the clinical treatment of different muscle growth disorders, including some muscular dystrophies (13, 16), and has the potential to significantly enhance the production of meat animal products as well (4, 6, 7, 10, 12). However, myostatin's functions are poorly defined in nonmammalian vertebrates and may be quite different in the fishes where multiple gene copies are differentially expressed in many tissues (5, 8, 1315). This is in stark contrast to mammals where myostatin expression is limited primarily to skeletal muscle and suggests that its functions in fish may be as diverse as its expression pattern. Thus, animal scientists and comparative biologists alike are greatly interested in extrapolating information between different animal models. This has proven quite difficult, however, as the current nomenclature for members of this transforming growth factor-β subfamily, which also includes growth differentiating factor (GDF)-11, is often confusing and sometimes problematic.

Myostatin is also known as GDF-8, although the former name is clearly the most popular. PubMed searches using the three known aliases (myostatin, GDF-8, or GDF8) and different Boolean conjunctions indicate that “myostatin” is more commonly used (Table 1). The exclusive use of myostatin occurred almost three times more often than did “GDF-8” (88 vs. 31 references), although both names were usually used in most articles since all field searches produced similar results. However, more restrictive searches of only title words indicate that myostatin is overwhelmingly favored (245 vs. 6 references), and thus GDF-8 is used primarily as an alias or keyword within the article. A disconnect therefore exists between researchers in the field and managers of public genome databases and some commercial institutes, including HUGO, GeneCards, and Jackson Laboratories, which prefer GDF-8 and include myostatin or its abbreviation, MSTN, as the alias. Describing the different orthologs and paralogs common to the fishes is particularly confusing because genes have been named according to the order by which they were identified within a specific species or by their tissue-specific expression pattern. Many were amended inconsistently with Roman or Arabic numbers or even with letters and in a manner independent of the phylogenetic relationships among the different gene copies. Extrapolating functional significance from studies that utilize different species is therefore exceedingly difficult, because genes with the same name are not necessarily orthologous.

The relatively recent phylogenetic analysis of vertebrate myostatin and GDF-11 homologs (summarized in Fig. 1) identified several gene duplication events that occurred throughout the evolution of this gene family (8). The first event separated the myostatin and GDF-11 lineages, and a second event produced two distinct myostatin paralogs in ray-finned fishes. Each paralog was subsequently duplicated once again before the radiation of the salmonids, likely due to tetraploidization, which produced two additional genes in this family. Barring losses within specific taxa, all modern bony fish should therefore possess a minimum of two myostatin genes, although salmonids in particular should have four. Indeed, four genes have recently been identified in the rainbow trout (GenBank accession nos. DQ136028, DQ138300, DQ138301, and DQ177320), which validates the hypothesis created by the phylogenetic analysis.

We propose a revised nomenclature that uses myostatin over GDF-8 and is based solely on the phylogenetic analysis (Fig. 1) (8). The nomenclature includes MSTN-1 and -2 genes in most fish species and MSTN-1a, -1b, -2a, and -2b specifically in the salmonids. This system was devised to minimize the number of name changes, to provide a guide for the naming of newly identified genes, and to facilitate the exchange of information. It is also consistent with the order by which the primary fish clades were identified. The consensus use of this simple, objective, and, most importantly, evolutionarily justified nomenclature will therefore end much confusion in the field and will assist in future discoveries (13, 16).

Fig. 1.

Phylogenetic distribution of different vertebrate myostatin (MSTN) homologs. The phylogenetic tree (left) of representative sequences was constructed from previous Maximum Likelihood and Bayesian Inference analyses (8). The 2 teleost fish paralogs, MSTN-1 and -2, are shaded. Within each clade are the additional salmonid paralogs; MSTN-1a and -1b are indicated by reverse shading and MSTN-2a and -2b with bold font. The specific GenBank accession numbers for cDNA sequences, the revised names, and the former aliases for each teleost homolog are included in the table.

View this table:
Table 1.

PubMed search results using different myostatin aliases