| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Laboratorio di Diagnosi Genetica,2 Unità Operativa Complessa di Pediatria e Genetica Medica, and3 Unità Operativa di Neurologia per il Ritardo Mentale, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Oasi Maria SS, Troina, Italy;4 Consultorio Genetico APSS, Trento, Italy;5 Department of Medical Genetics, Yorkhill Hospitals, Glasgow, Scotland, United Kingdom
aaddress correspondence to this author at: IRCCS Oasi Maria SS, via Conte Ruggero 73, 94018 Troina, Italy; fax 39-0935-653327, e-mail mfichera{at}oasi.en.it
Coffin–Lowry syndrome (CLS; MIM #303600) is characterized by learning difficulties and dysmorphic traits in male patients and in some female carriers of the X-chromosome–linked gene mutation. The dysmorphic traits, skeletal abnormalities, and other clinical findings have been described (1). Mutations of the RSK2 gene (also called RPS6KA3, MIM *300075), mapping to Xp22.2, are found in the disease (2). The gene encodes a 740-amino acid protein member of the 90-kDa ribosomal S6 serine/threonine kinase family (3). In humans, the RSK family includes 4 growth factor–regulated members (RSK1 to -4) produced in all examined tissues and regions of the brain (4). The highly conserved feature of all these proteins is the presence of 2 nonidentical kinase catalytic domains. The N-terminal kinase domain (amino acids 68–323 in RSK2) is responsible for phosphotransferase activity toward substrates, whereas the C-terminal kinase domain (amino acids 422–675 in RSK2) is necessary for enzymatic activation of the N-terminal domain. RSKs are activated through direct phosphorylation by the mitogen-activated protein kinases/extracellular signal-regulated kinases (MAPK/ERK) in response to insulin and growth factors, oncogenic events, and ultraviolet irradiation. RSKs have been implicated in the stimulation of cell proliferation and differentiation, in the cellular stress response, and in apoptosis (5). To date, only a few RSK2-specific physiologic substrates have been described: the transcription factor CREB, histone H3, STAT3 (6), ATF4(7), and p53(8).
According to the most recent data (http://alsace.u-strasbg.fr/chimbio/diag/coffin/index.html), mutations in the RSK2 gene occur within all 22 exons, except exon 2. Approximately one third of the RSK2 gene mutations are missense changes, the other two thirds lead to premature translation termination. The proportion of de novo mutations is unusually high compared with other X-linked disorders, with most cases (80%) being sporadic (1). The 2 largest studies on the molecular diagnosis of CLS used single-strand conformational polymorphism analysis (3)(9). The large size of the transcript (a 2223-bp open reading frame) and the distribution of mutations all along its length require a cost-effective screening technique. We propose a highly sensitive and rapid approach to mutation detection based on denaturing HPLC (DHPLC) (10) to screen for RSK2 mutations.
We studied 16 (10 male, 6 female) unrelated Italian and British individuals. Among them, 9 showed a typical CLS phenotype, whereas the remaining 7 exhibited only some of the characteristic features of the syndrome. As described below, a CLS phenotype was also observed in mothers of 3 of the typical CLS patients, but none of the CLS patients had other affected relatives. Genomic DNA was isolated from leukocytes in peripheral blood by salting out procedures after receipt of informed written consent from the parents. DNA samples from 150 unrelated healthy females (300 chromosomes) of Italian and British origin were screened to determine whether the novel mutations identified in this study were present in a normal population.
PCR primers (see Table S1 in the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol51/issue12) were designed by the software Vector NTI AdvanceTM (Invitrogen) to amplify all 22 exons and flanking intronic sequences of the RSK2 gene (Entrez Gene ID 6197). PCR reactions (total volume, 50 µL) contained 200 ng of genomic DNA, 1 U of AmpliTaq GoldTM polymerase in 1x buffer (both from Applied Biosystems), 2.5 mM MgCl2, 0.2 mM each of the deoxynucleotide triphosphates, and 1 µM each of the primers. The PCR conditions included an initial denaturation step at 94 °C for 5 min, followed by 30 cycles at 94 °C for 40 s, 56 °C (63 °C for exon 1 with 1.5 M betaine) for 40 s, and 72 °C for 40 s, and a final extension step of 72 °C for 7 min. Heteroduplex formation for the analysis of DNA from male patients was induced by mixing equimolar amounts of PCR products from the patients and a sequence-confirmed wild-type control, denaturing at 95 °C for 5 min, and gradual reannealing from 95 °C to 25 °C over a period of 30 min.
DHPLC analysis was carried out on a WAVETM 3500HT System (Transgenomic). Crude PCR products (8 µL) were eluted at a flow rate of 1.5 mL/min (2 min per elution), with a 5% linear acetonitrile gradient. WAVE OptimizedTM buffers A (0.1 mol/L triethylammonium acetate) and B (0.1 mol/L triethylammonium acetate containing 250 mL/L acetonitrile) from Transgenomic were used to ensure highly reproducible retention times. Resolution temperatures for each amplicon assessment were tested within a 3 °C window above and below the melting temperature suggested by the NavigatorTM software (Transgenomic), and finally, 2 or 3 analysis temperatures were established (see Table S1 in the online Data Supplement).
Patients’ chromatograms were compared with corresponding wild-type controls included in each run. Heterozygous profiles were detected as distinct elution peaks from unique homozygous wild-type peaks.
Fresh PCR products with a heterozygous pattern were purified with the QIAquick PCR purification method (Qiagen) and were bidirectionally sequenced on an ABI 310 Genetic Analyzer® (Applied Biosystems) using the same primers of DHPLC analysis and the BigDye® chemistry (Applied Biosystems).
A total of 13 abnormal elution profiles were detected, 7 of which were considered pathogenic (see Fig. S1 in the online Data Supplement) and were found in typical CLS patients.
In 2 female patients, we detected the previously reported (3) mutations p.R110X and p.R514X. Five anomalous peaks were produced by novel mutations: c.683T>G (p.F228C), c.1402_1404delCTT (p.L468del), IVS17-5A>G (Fig. 1
), and c.2180C>G (p.A727G) in males and c.1086delT (p.A363QfsX424) in 1 female. None of the new mutations were found in the control population.
|
Among the other 6 heterozygous profiles, 2 revealed already reported polymorphic variants, the nonsynonymous change c.113T>G (p.I38S) and the synonymous single-nucleotide polymorphism c.798C>A (rs12009120), and 4 were unreported intronic variations, IVS2 + 21C>T, IVS7-67A>C, IVS9 + 28A>G, and IVS17-24C>A. All of the novel sequence polymorphisms were also found in the control population.
The IVS17-5A>G mutation creates a novel acceptor site, as predicted by the online software Spliceview (http://www.itba.mi.cnr.it/webgene), 4 nucleotides before the correct site with a predicted value stronger than that of the normal site (a score of 82 vs 78). If used, this abnormal site affects the correct splicing of intron 17, producing a TTAG insertion before exon 18, hence introducing a frameshift at codon 535 with a premature termination signal at position 544. Testing for this hypothesis, we investigated RSK2 expression in leukocyte RNA from this patient, using a pair of primers (forward primer, 5'-GAACGAGAGGCCAGTGCTGT-3'; reverse primer, 5'-CGCTCTCAGCTGTTTTGCAA-3') specific for the exon17/18 junction. The reverse transcription-PCR product was then cloned; among 20 sequenced clones, 18 revealed the predicted abnormal fragment (data not shown).
The p.F228C and p.A727G missense mutations are located, respectively, in the N-terminal kinase domain and in the ERK docking site and involve residues highly conserved during evolution in the RSK family from invertebrate to humans (data not shown). Surprisingly, these mutations were associated with mild (p.F228C) or moderate (p.A727G) mental retardation in affected males. The mothers of these 2 patients are heterozygous carriers of these mutations, and their faces and hands are suggestive of the CLS phenotype, but they demonstrate no cognitive impairment. These mutations were not observed in other healthy relatives of these patients. It is likely that these alterations, as already reported for other RSK2 missense mutations (11)(12), do not entirely impair the protein function.
The third familial case was the female carrying the p.R514X mutation, who has a carrier mother showing a phenotype very similar to that of her daughter. Paroxysmal movement disorder is a particular problem for patients with the p.R110X mutation, as was described recently (13). Although p.R110X is one of the few recurrent mutations of this gene, we were unable to find other clinical descriptions of female patients carrying this mutation, and the movement disorder in CLS, although common, is not well studied.
No clear hot spot mutations were found, thus highlighting the functional relevance of both protein kinase domains. In patients negative for RSK2 mutations by DHPLC analysis, direct sequencing of the entire gene failed to detect any variations. Moreover, no false-positive results were reported, each abnormal chromatogram containing 1 mutation or polymorphism.
The detection rate in our patients was 7 of 16 (44%), and despite the small number of patients, we believe that it represents a significant increase in the sensitivity of RSK2 mutational screening compared with the recent literature. Among patients who were negative on RSK2 mutation analysis, only 2 exhibited a typical CLS phenotype. As suggested previously (14), genetic heterogeneity or defects in regions not investigated by this assay can account for the failure in detecting RSK2 mutations in those patients.
Including the time to set up the PCR reaction and to perform the DHPLC, analysis time was 
6 h to screen the entire coding region of the RSK2 gene, thus providing a high-throughput alternative to single-strand conformational polymorphism–based analysis for molecular prescreening in CLS patients.
Acknowledgments
We gratefully acknowledge financial support from the Italian Ministry of Health.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
