1. ¹Eugenio Medea Scientific Institute, Bosisio Parini, Lecco, Italy
2. ²Department of Experimental and Diagnostic Medicine, Medical Genetic Section, University of Ferrara, Ferrara, Italy
3. ³UO Genetica Medica, AUSL Imola, Imola, Italy
4. ⁴The Wellcome Trust Sanger Institute, Hinxton, Cambridge, UK
5. ⁵European Bioinformatics Institute (EBI) Wellcome Trust Genome Campus, Hinxton, Cambridge, UK
6. ⁶Biologia Generale e Genetica Medica, Università di Pavia, Pavia, Italy
7. ⁷Fondazione IRCCS C Mondino, Pavia, Italy

Correspondence: Dr MC Bonaglia, IRCCS Eugenio Medea, Via Don Luigi Monza, 20, 23842 Bosisio Parini, Lecco, Italy. Tel: +39 031 877 111; Fax: +39 031 877 499; E-mail: clara.bonaglia@bp.lnf.it

⁸These authors contributed equally to this study.

Received 22 July 2008; Revised 11 September 2008; Accepted 12 September 2008; Published online 15 October 2008.

Abstract

Although 22q terminal deletions are well documented, very few patients with mosaicism have been reported. We describe two new cases with mosaic 22q13.2-qter deletion, detected by karyotype analysis, showing the neurological phenotype of 22q13.3 deletion syndrome. Case 1 represents an exceptional case of mosaicism for maternal 22q13.2-qter deletion (45% of cells) and 22q13.2-qter paternal segmental isodisomy (55% of cells). This complex situation was suspected because cytogenetic, FISH and array-CGH analyses showed the presence of an 8.8 Mb mosaic 22q13.2-qter deletion, whereas microsatellite marker analysis was consistent with maternal deletion without any evidence of mosaic deletion. Molecular analysis led to the definition of very close, but not coincident, deletion and uniparental disomy (UPD) break points. Furthermore, we demonstrated that the segmental UPD arose by gene conversion in the same region. In Case 2, mosaicism for a paternal 8.9 Mb 22q13.2-qter deletion (73% of cells) was detected. In both patients, the level of mosaicism was also verified in saliva samples. We propose possible causative mechanisms for both rearrangements. Although the size of the deletions was quite similar, the phenotype was more severe in Case 2 than in Case 1. As maternal UPD 22 has not been generally associated with any defects and as the size of the deletion is very similar in the two cases, phenotype severity is likely to depend entirely on the degree of mosaicism in each individual.