Statistics from Altmetric.com
Recently, missense mutations in the PTPN11 gene on chromosome 12q24.1 have been identified as the cause of Noonan’s syndrome in 45% of familial and sporadic cases,4,5 indicating genetic heterogeneity within the syndrome. In the study by Tartaglia et al,5 there was a family in which three members had features of Noonan’s syndrome; two of these had incidental mandibular giant cell lesions.3 All three members were found to have a PTPN11 mutation known to cosegregate with the Noonan phenotype. This mutation, an A→G transition at position 923 in exon 8, predicting an Asn308Ser substitution within the PTP domain, was identified in an unrelated kindred with classical Noonan’s syndrome. No other patients with NL/MGCLS had been evaluated for the PTPN11 mutation.
Cherubism is caused by a missense mutation in the coding region of the SH3BP2 gene on chromosome 4p16.3.6 In the study by Ueki et al,6 12 of 15 families showed point mutations in the SH3 binding protein, SH3BP2. All seven mutations identified were on exon 9 and affected three amino acids by substitution within a six amino acid sequence. A second locus or gene has not been identified.
We present the phenotype of three sporadic cases of NS/MGCLS and the results of mutation analysis of the PTPN11 and SH3BP2 genes.
The clinical features of the three patients are summarised in table 2.
All patients were enrolled in the National Institutes of Health IRB approved protocols and written informed consent was obtained. The three patients were diagnosed with NL/MGCLS from their clinical findings. There was no family history of cherubism, Noonan’s syndrome, congenital heart disease, or consanguinity.
G banded karyotyping was undertaken using standard techniques. The entire PTPN11 coding region was screened, as previously reported.5 Briefly, unpurified polymerase chain reaction (PCR) products were analysed by denaturing high performance liquid chromatography (DHPLC), using the Wave DNA fragment analysis system (Transgenomics, Omaha, Nebraska, USA) at column temperatures recommended by the WaveMaker version 4.1.31 software (Transgenomics). Heterozygous templates with previously identified mutations were used as positive controls for all exons. Amplimers having abnormal denaturing profiles were purified (Microcon PCR, Millipore, Bedford, Massachusetts, USA) and sequenced bidirectionally using the ABI BigDye terminator sequencing kit v.3.1 (Applied Biosystems, Foster City, California, USA) and an ABI Prism 310 genetic analyser (Applied Biosystems). Sequencing results were analysed using the Sequencing Analysis v.3.6.1 (Applied Biosystems) and AutoAssembler v.2.1 software packages (Applied Biosystems). Mutation analysis of the SH3BP2 gene was carried out as previously published.6
Noonan-like/multiple giant cell lesion syndrome (NL/MGCLS) has clinical similarities with Noonan’s syndrome and cherubism. It is unclear whether it is a distinct entity or a variant of Noonan’s syndrome or cherubism.
Three unrelated patients with NL/MGCLS were characterised, two of whom were found to have mutations in the PTPN11 gene, the mutation found in 45% of patients with Noonan’s syndrome. None of the patients had a mutation of the SH3BP2 gene known to cause cherubism.
Giant cell lesions are likely to be a part of the spectrum of findings in Noonan’s syndrome and not a distinct entity.
G banded karyotype analysis was normal in all three patients at a 550 band resolution. The entire coding sequences of the PTPN11 and SH3BP2 genes were screened by DHPLC analysis and direct sequencing. PTPN11 mutation screening identified different heterozygous missense mutations in patients 1 and 3 (fig 3). The former was an A→C transition at position 317 in exon 3, resulting in the Asp106Ala substitution within the N-SH2/C-SH2 linker. Patient 3 showed a T→C transition at position 853 in exon 7, predicting a Phe285Leu substitution within the PTP domain. Both mutations were de novo (fig 3) and had been documented previously among individuals with Noonan’s syndrome.5 No mutation within the SH3BP2 gene was identified in any of the patients.
We report three unrelated patients with NL/MGCLS, two with PTPN11 mutations and none with SH3BP2 gene changes. The presence of these mutations supports the previous assertion that NL/MGCLS is an extreme phenotype of Noonan’s syndrome. The failure to detect a PTPN11 mutation in the third subject suggests that NL/MGCLS, like Noonan’s syndrome, is genetically heterogeneous. While the promoter and enhancer regions were not examined, the nature and functional effects of the PTPN11 lesions observed in Noonan’s syndrome and NL/MGCLS make the possibility of a mutation in those non-coding regions highly unlikely.
PTPN11 encodes the non-receptor protein tyrosine phosphatase, SHP-2 (src homology region 2-domain phosphatase-2).7 SHP-2 is essential in multiple intracellular signal transduction pathways that affect but are not limited to mesodermal patterning and limb development,8,9 epidermal growth factor receptor signalling,10 and cardiac semilunar valvogenesis.11 The highly conserved functional domains of the SHP-2 protein comprise two tandemly arranged SH2 domains at the N terminus (N-SH2 and C-SH2) followed by a catalytic protein tyrosine phosphatase (PTP) domain, and a carboxy-terminal tail.12,13 In the inactive conformation of this structure, N-SH2 and PTP interact through multiple hydrogen bonds and polar interactions blocking the PTP active site.14–17
In the study by Tartaglia et al,5 most of the missense mutations affected the amino acids located in the N-SH2 and PTP functional domains, with the majority of these mutations directly involved in or located near the interacting region. This distribution suggests that the pathogenic mechanism involves an altered N-SH2/PTP interaction that destabilises the inactive conformation without altering SHP-2 catalytic capability. Molecular modelling and the first functional data support a model in which PTPN11 mutations upregulate SHP-2 physiological activation by impairing the switch between the active and inactive conformation, favouring a shift in the equilibrium toward the active conformation and a gain of function.4,18,19
There does not appear to be a mutation that is consistent with the presence of giant cell lesions in these patients. The identification of three PTPN11 mutations (exon 3, Asp106Ala in patient 1; exon 7, Phe285Leu in patient 3; exon 8, Asn308Ser5) in cases of Noonan’s syndrome with and without multiple giant cell lesions suggests that additional events may contribute to such phenotypic heterogeneity. This may include second hits in the same gene or in genes coding for signalling molecules with a role in transduction pathways in which SHP-2 is involved. Though the actual mechanism of SHP-2 in giant cell development and lesion formation is unclear, there is evidence that it is important in myeloid cell proliferation and differentiation. Gain of function somatic mutations of PTPN11 have been identified in patients with juvenile myleomonocytic leukaemia with or without Noonan’s syndrome, myelodysplastic syndromes, and acute myeloid leukaemia, conditions in which malignant transformation has affected the myeloid precursor cells.18
All three of the patients in this study had pulmonary stenosis confirmed by echocardiography. In a review of published reports and including the three study patients, there are 24 reported cases of NL/MGCLS, of which 17 (70.8%) had pulmonary stenosis.1,3,20–24 In Noonan’s syndrome, over 80% of the patients have a cardiovascular abnormality, pulmonary stenosis being the most common defect.25,26 The high prevalence of pulmonary stenosis within the NL/MGCLS population suggests that PTPN11 will be the dominant mutated gene in this syndrome. In the study by Tartaglia et al,5 pulmonary stenosis was the most common cardiac defect and in the affected cases 70.6% had a mutation in the PTPN11 gene (p = 0.008). The frequent presence of cardiac defects in NL/MGCLS decreases the likelihood that it is a separate entity from Noonan’s syndrome.
The three patients were all found to have low bone density compared with age matched control data. To our knowledge, low bone density has not been described previously in either NL/MGCLS, Noonan’s syndrome, or cherubism. However, generalised hypomineralisation was mentioned in the case report by Cohen and Gorlin1 of a patient with features of Noonan’s syndrome and giant cell lesions. The clinical history of our three patients did not reveal an increased fracture rate. Lesions in the craniofacial region appear to be caused primarily by expansion of the cells of the bone marrow stromal compartment, and low bone mass may reflect the effects of the mutation on these cells in the appendicular bones.
In these sporadic cases of NL/MGCLS, the giant cell lesions are identical to those of cherubism by histology and clinical presentation. However, the mutations in the cherubism gene, SH3BP2, are absent in these patients. As mutations in the Noonan’s syndrome gene are known at this time, and were found in two of the three patients, it is likely that the giant cell lesions are a part of the spectrum of findings in Noonan’s syndrome and not a distinct entity. The diagnosis of Noonan’s syndrome continues to expand, and its clinical features now include giant cell lesions. However, it is unclear what additional pathogenic factors result in the formation of these giant cell lesions.
This work was supported in part by NIH grants HD01294, HL074728, and HL71207 to BDG; and Telethon-Italy grant GGP04172 to MT.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.