The Tale of Fop, Noggin and Myristoylation: No Data, No Proof!

It appears to us that a recent letter to the editor of Genetic Counseling by Lucotte et al. (5) reports on patients that had been already published elsewhere (1,7, 9). These publications were discussed controversially by letters to the editor (10, 11). The scientific community needs to know if these patient data are new or if the authors publish the same story in several journals to give the impression that many so-called FOP patients exist that have mutations in NOG. Considering the intense discussion their first papers evoked, it appears awkward to us that the authors do not even try to convince the scientific community that their findings are correct by discussing the previous controversial statements or by providing more details to prove their point; e.g. publish patient documentation, use techniques for diagnosis of NOG mutations as published by other groups (2, 4). Besides these concerns, there are some more serious flaws concerning the authors' conclusions. Firstly, the authors do not follow the standard nomenclature of mutations - what they refer to as 276A should be c.276G>A (p.Gly93Gly); which means that this exchange is a silent mutation and is therefore unlikely to be disease-causing. Secondly, the authors point out that all point mutations affect the polyGly stretch of the NOG protein (amino acid 89-97 in the human NOG sequence NP005441). Lucotte et al. claim that this polyGly stretch represents a myristoylation site that is altered by the mutations. They support this hypothesis by citing own work (6). This reference insinuates to the reader that it was already functionally verified by additional experimental data. However, when checking the reference it became apparent that it refers to an oral communication at the Second International Workshop on the Genetics of Bone Metabolism and Disease in Davos Congress Centre, Switzerland from February 15th-18th 2003, organized by the European Calcified Tissue Society, published in abstract form at the conference in Cale Tissues Intern. Since 2003 there were no further publications by this group providing any new data. In our opinion it is very unlikely that this polyGly stretch is a true myristoylation site. Myristoylation is a common lipid modification of intracellular proteins, which fully depends on an N-terminal Gly-residue. The N-terminal GIy usually follows the initiation Met or becomes available after limited proteolysis (8). NOG does not represent an intracellular protein; moreover it is not processed at the polyGly stretch and therefore it is no substrate for myristoylation. This conclusion is also supported by processing the human NOG protein sequence via an open access tool to predict myristoylation sites (MYRbase: http://mendel.imp.ac.at/myristate/SUPLpredictor.htm). Instead, it is very likely that the polyGly stretch of NOG, which is only present in mammalian NOG isoforms, is a highly flexible structure and becomes therefore not visible in the crystal analysis (3). The polyGly stretch is not important for the inhibitory effect of NOG on BMPs as shown by the chicken NOG, which is able to inhibit humans BMPs (own observation) in spite of the fact that it lacks this polyGly stretch. It appears to us as if the myristoylation theory of NOG represents rather pure speculation than solid science, since experimental data are missing completely. …