No mutation in the gene for Noonan syndrome, PTPN11, in 18 patients with Costello syndrome.

Abstract

Costello syndrome (OMIM 218040; faciocutaneoskeletal syndrome; FCS syndrome) is a rare disorder characterized by postnatal growth deficiency, mental retardation, “coarse” face, loose skin of the neck, hands, and feet, cardiomyopathy, and nasal papillomata [Costello, 1971, 1977]. To date, the causative gene for Costello syndrome is not known. Costello syndrome, cardiofaciocutaneous syndrome (CFC; MIM 115150), and Noonan syndrome (NS; MIM 163950) display some phenotypic similarities, prompting a controversy whether or not CFC, Noonan, and Costello syndromes represent different traits. Several authors have proposed that CFC and NS are variable manifestations of the same entity since both share considerable similarities [Fryer et al., 1991; Ward et al., 1994; Leichtman, 1996; Lorenzetti and Fryns, 1996; Legius et al., 1998]. Others have argued that Noonan syndrome and CFC are similar but distinct, specific entities, whose diagnosis is not interchangeable [Neri et al., 1991; Neri and Zollino, 1996]. Costello syndrome can be added to this group as a third disorder with clinical overlap. Wieczorek et al. [1997] compiled the clinical findings of CFC, Noonan, and Costello syndromes and discussed the difficulties of diagnosing each condition conclusively. They concluded that the ectodermal anomalies in CFC and Costello syndrome may help to distinguish clearly between the three disorders. The issue of differential diagnosis will be solved in part following identification of the causative gene for at least one entity and the absence of disease-relevant mutations in the corresponding gene in patients with one or both of the remaining two conditions. Indeed, no mutation was found in one of the genes for NS, PTPN11, in 28 cases with CFC [Ion et al., 2002] suggesting that Noonan syndrome and CFC represent separate conditions. Yet, the same question regarding Noonan and Costello syndromes is still unanswered. PTPN11 encodes the protein tyrosine phosphatase SHP-2 [Freeman et al., 1993]. Mutations in PTPN11 account for 45–50% of cases of sporadic or familial NS [Tartaglia et al., 2001, 2002]. Co-segregation of a PTPN11 mutation in a family with the Noonan-like/multiple giant-cell lesion syndrome (MIM 163955) has also been demonstrated [Tartaglia et al., 2002]. Moreover, two groups have shown independently that multiple-lentigines (ML)/LEOPARD syndrome (MIM 151100) is also caused by mutations in PTPN11 [Digilio et al., 2002; Legius et al., 2002]. Although both Noonan-like/multiple giant-cell lesion and multiple-lentigines (ML)/LEOPARD syndromes were formerly considered distinct disorders, current data demonstrate that (in some cases) these syndromes and NS are allelic disorders. These results also imply that mutation screening of PTPN11 for other NS-like conditions is recommended. Thus, we considered PTPN11 a candidate gene for Costello syndrome. We collected DNA samples of 18 patients clinically diagnosed as Costello syndrome. Detailed phenotypic data were available for 14 patients and are summarized in Table I. All patients show the typical facial appearance with coarse face, thick lips, low set ears, short neck, and full cheeks. Pulmonic stenosis and hypertrophic cardiomyopathy are two frequent cardiac features of Costello syndrome [Lin et al., 2002]. Indeed, of the 11 patients with cardiac defects, six and seven have pulmonic stenosis and hypertrophic cardiomyopathy, respectively, with two patients presenting with both conditions. For mutation screening, the 15 coding exons of PTPN11 and flanking intronic sequences were amplified by PCR [Tartaglia et al., 2001] using genomic DNA of each patient as template. The amplicons were further analyzed by single strand conformational polymorphism (SSCP) as described previously [Bunge et al., 1996] and fragments that showed mobility shifts of single strands and/or heteroduplex formation were sequenced. We did not detect pathogenic mutations in any patient. In one patient, we observed the sequence variant IVS9-9C/A that has already been described as a polymorphism [Tartaglia et al., 2002]. In this study, we have excluded the NS gene PTPN11 as the major causative gene for Costello syndrome. Clearly, we cannot rule out that molecular lesions might be present in PTPN11 in Costello patients that escaped detection by our mutation screening, such as large heterozygous deletions, duplications, inversions, changes in the 5′ and 3′ UTR, or promoter defects. Our conclusion that PTPN11 is not likely involved in the pathogenesis of Costello syndrome suggests that Costello syndrome and NS may still be considered two distinct entities. Further studies are required to determine the molecular basis for both Costello syndrome and CFC and for the NS patients with no PTPN11 mutation. Once a gene for CFC and/or Costello syndrome has been identified, it will be of great interest to investigate whether these syndromes are allelic disorders. The results summarized here are part of the M.D. thesis of Birte Tröger at the University Hospital Hamburg-Eppendorf. We gratefully acknowledge the participation of the patients with Costello syndrome and their families as well as the Costello Syndrome Support Group.