Portal de Periódicos da CAPES

3q29 Microdeletion Syndrome: Clinical and Molecular Characterization of a New Syndrome

2005; Elsevier BV; Volume: 77; Issue: 1 Linguagem: Inglês

10.1086/431653

1537-6605

Lionel Willatt, James J. Cox, John Barber, Elisabet Dachs Cabanas, Amanda Collins, Dian Donnai, David Fitzpatrick, Eddy Maher, Howard Martin, Josep Parnau, Lesley Pindar, Jacqueline Ramsay, Charles Shaw‐Smith, Erik A. Sistermans, Michael Tettenborn, Dorothy Trump, Bert B.A. de Vries, Kate Walker, F. Lucy Raymond,

RNA regulation and disease

We report the identification of six patients with 3q29 microdeletion syndrome. The clinical phenotype is variable despite an almost identical deletion size. The phenotype includes mild-to-moderate mental retardation, with only slightly dysmorphic facial features that are similar in most patients: a long and narrow face, short philtrum, and high nasal bridge. Autism, gait ataxia, chest-wall deformity, and long and tapering fingers were noted in at least two of six patients. Additional features—including microcephaly, cleft lip and palate, horseshoe kidney and hypospadias, ligamentous laxity, recurrent middle ear infections, and abnormal pigmentation—were observed, but each feature was only found once, in a single patient. The microdeletion is ∼1.5 Mb in length, with molecular boundaries mapping within the same or adjacent bacterial artificial chromosome (BAC) clones at either end of the deletion in all patients. The deletion encompasses 22 genes, including PAK2 and DLG1, which are autosomal homologues of two known X-linked mental retardation genes, PAK3 and DLG3. The presence of two nearly identical low-copy repeat sequences in BAC clones on each side of the deletion breakpoint suggests that nonallelic homologous recombination is the likely mechanism of disease causation in this syndrome. We report the identification of six patients with 3q29 microdeletion syndrome. The clinical phenotype is variable despite an almost identical deletion size. The phenotype includes mild-to-moderate mental retardation, with only slightly dysmorphic facial features that are similar in most patients: a long and narrow face, short philtrum, and high nasal bridge. Autism, gait ataxia, chest-wall deformity, and long and tapering fingers were noted in at least two of six patients. Additional features—including microcephaly, cleft lip and palate, horseshoe kidney and hypospadias, ligamentous laxity, recurrent middle ear infections, and abnormal pigmentation—were observed, but each feature was only found once, in a single patient. The microdeletion is ∼1.5 Mb in length, with molecular boundaries mapping within the same or adjacent bacterial artificial chromosome (BAC) clones at either end of the deletion in all patients. The deletion encompasses 22 genes, including PAK2 and DLG1, which are autosomal homologues of two known X-linked mental retardation genes, PAK3 and DLG3. The presence of two nearly identical low-copy repeat sequences in BAC clones on each side of the deletion breakpoint suggests that nonallelic homologous recombination is the likely mechanism of disease causation in this syndrome. The history of detecting microdeletion syndromes started with the recognition of discrete syndromic phenotypes associated with mental retardation, such as Prader-Willi, Miller-Dieker, Angelman, and Williams syndromes. Patients with similar phenotypes were then found to have similar submicroscopic deletions, in discrete genomic regions, that then defined the condition (Ledbetter et al. Ledbetter et al., 1981Ledbetter DH Riccardi VM Airhart SD Strobel RJ Keenan BS Crawford JD Deletions of chromosome 15 as a cause of the Prader-Willi syndrome.N Engl J Med. 1981; 304: 325-329Crossref PubMed Scopus (456) Google Scholar; Schwartz et al. Schwartz et al., 1988Schwartz CE Johnson JP Holycross B Mandeville TM Sears TS Graul EA Carey JC Schroer RJ Phelan MC Szollar J Flannery DB Stevenson RE Detection of submicroscopic deletions in band 17p13 in patients with the Miller-Dieker syndrome.Am J Hum Genet. 1988; 43: 597-604PubMed Google Scholar; Knoll et al. Knoll et al., 1989Knoll JH Nicholls RD Magenis RE Graham Jr, JM Lalande M Latt SA Angelman and Prader-Willi syndromes share a common chromosome 15 deletion but differ in parental origin of the deletion.Am J Med Genet. 1989; 32: 285-290Crossref PubMed Scopus (393) Google Scholar; Ewart et al. Ewart et al., 1993Ewart AK Morris CA Atkinson D Jin W Sternes K Spallone P Stock AD Leppert M Keating MT Hemizygosity at the elastin locus in a developmental disorder, Williams syndrome.Nat Genet. 1993; 5: 11-16Crossref PubMed Scopus (896) Google Scholar). In 1995, Flint et al. developed a strategy to screen for the abnormal inheritance of some subtelomeric DNA polymorphisms in individuals who had mental retardation alone and no associated clinical syndrome (Flint et al. Flint et al., 1995Flint J Wilkie AO Buckle VJ Winter RM Holland AJ McDermid HE The detection of subtelomeric chromosomal rearrangements in idiopathic mental retardation.Nat Genet. 1995; 9: 132-140Crossref PubMed Scopus (448) Google Scholar). Initially, 3 of 99 patients each had a single deletion at one of the telomeres. Since then, the technology has been developed to provide screening of all telomeres as a clinical service for suitably selected patients with mental retardation and dysmorphic features. The range of diagnostic yield is ∼6%–11% (Knight et al. Knight et al., 1999Knight SJ Regan R Nicod A Horsley SW Kearney L Homfray T Winter RM Bolton P Flint J Subtle chromosomal rearrangements in children with unexplained mental retardation.Lancet. 1999; 354: 1676-1681Abstract Full Text Full Text PDF PubMed Scopus (412) Google Scholar; Slavotinek et al. Slavotinek et al., 1999Slavotinek A Rosenberg M Knight S Gaunt L Fergusson W Killoran C Clayton-Smith J Kingston H Campbell RH Flint J Donnai D Biesecker L Screening for submicroscopic chromosome rearrangements in children with idiopathic mental retardation using microsatellite markers for the chromosome telomeres.J Med Genet. 1999; 36: 405-411PubMed Google Scholar; Flint and Knight Flint and Knight, 2003Flint J Knight S The use of telomere probes to investigate submicroscopic rearrangements associated with mental retardation.Curr Opin Genet Dev. 2003; 13: 310-316Crossref PubMed Scopus (123) Google Scholar; Koolen et al. Koolen et al., 2004Koolen DA Nillesen WM Versteeg MH Merkx GF Knoers NV Kets M Vermeer S van Ravenswaaij CM de Kovel CG Brunner HG Smeets D de Vries BB Sistermans EA Screening for subtelomeric rearrangements in 210 patients with unexplained mental retardation using multiplex ligation dependent probe amplification (MLPA).J Med Genet. 2004; 41: 892-899Crossref PubMed Scopus (135) Google Scholar). The frequency of specific chromosome submicroscopic deletions or duplications reported in the literature varies from single case reports to >50 cases reported (De Vries et al. De Vries et al., 2003De Vries BB Winter R Schinzel A van Ravenswaaij-Arts C Telomeres: a diagnosis at the end of the chromosomes.J Med Genet. 2003; 40: 385-398Crossref PubMed Scopus (197) Google Scholar). Conditions such as 1p36 or 2q37.3 microdeletions have been reported frequently, and these conditions now assume the status of recognizable syndromes, since common clinical features have emerged through the collection of patients with the same deletion (Shapira et al. Shapira et al., 1997Shapira SK McCaskill C Northrup H Spikes AS Elder FF Sutton VR Korenberg JR Greenberg F Shaffer LG Chromosome 1p36 deletions: the clinical phenotype and molecular characterization of a common newly delineated syndrome.Am J Hum Genet. 1997; 61: 642-650Abstract Full Text PDF PubMed Scopus (204) Google Scholar; Heilstedt et al. Heilstedt et al., 2003Heilstedt HA Ballif BC Howard LA Lewis RA Stal S Kashork CD Bacino CA Shapira SK Shaffer LG Physical map of 1p36, placement of breakpoints in monosomy 1p36, and clinical characterization of the syndrome.Am J Hum Genet. 2003; 72: 1200-1212Abstract Full Text Full Text PDF PubMed Scopus (211) Google Scholar; Aldred et al. Aldred et al., 2004Aldred MA Sanford RO Thomas NS Barrow MA Wilson LC Brueton LA Bonaglia MC Hennekam RC Eng C Dennis NR Trembath RC Molecular analysis of 20 patients with 2q37.3 monosomy: definition of minimum deletion intervals for key phenotypes.J Med Genet. 2004; 41: 433-439Crossref PubMed Scopus (77) Google Scholar). However, several of the subtelomeric deletions have been reported only rarely to date, and the delineation of the associated clinical phenotype is therefore much more difficult. We report here the identification of six patients with a 3q29 microdeletion. In five patients, the deletion arose de novo, and, for one patient, the parental origin could not be determined. The clinical phenotype is described in all six patients and is compared with that of a single case that was reported by Rossi et al. (Rossi et al., 2001Rossi E Piccini F Zollino M Neri G Caselli D Tenconi R Castellan C Carrozzo R Danesino C Zuffardi O Ragusa A Castiglia L Galesi O Greco D Romano C Pierluigi M Perfumo C Di Rocco M Faravelli F Dagna Bricarelli F Bonaglia M Bedeschi M Borgatti R Cryptic telomeric rearrangements in subjects with mental retardation associated with dysmorphism and congenital malformations.J Med Genet. 2001; 38: 417-420Crossref PubMed Google Scholar). The molecular boundaries of the recurrent 1.5-Mb deletion are defined, and nonallelic homologous recombination is proposed as a mechanism of deletion formation in these patients. Table 1 summarizes the clinical features of the six patients presented here and includes the few clinical details available on the patient analyzed in the study by Rossi et al. (Rossi et al., 2001Rossi E Piccini F Zollino M Neri G Caselli D Tenconi R Castellan C Carrozzo R Danesino C Zuffardi O Ragusa A Castiglia L Galesi O Greco D Romano C Pierluigi M Perfumo C Di Rocco M Faravelli F Dagna Bricarelli F Bonaglia M Bedeschi M Borgatti R Cryptic telomeric rearrangements in subjects with mental retardation associated with dysmorphism and congenital malformations.J Med Genet. 2001; 38: 417-420Crossref PubMed Google Scholar). Figure 1 illustrates the patients' facial features and hands. To summarize, this condition is not associated with any antenatal abnormalities, and the birth history was uneventful in all patients. Birth measurements were within the normal range, although microcephaly was noted in patient 6. The level of developmental delay often was not fully recognized until after the 1st year of life, except in patient 6, in whom failure to thrive was noted at 5 mo of age. Growth parameters in all patients were at or below the 50th percentile but remained in the normal range. Speech delay was a particular feature, and all children needed special educational provision because of more general intellectual difficulties, although one child was in a mainstream school with substituted lessons in a specialist learning support unit. Autism was a feature of the behavior of two of the patients. Patient 1's behavior was obsessive and repetitive and met the ICD-10 (International Classification of Diseases, World Health Organization) diagnostic criteria for autism spectrum disorders; patient 3 presented at 3 years of age with moderate-to-severe cognitive learning difficulties with autistic traits but was not diagnosed as having a recognized form of autism. His psychological presentation was complex at that stage and remained so. The dysmorphology in the children is variable and not striking, but there are some facial similarities, and each child also has certain unique features. Generally, the face is long and narrow, especially in patients 2 and 4. The philtrum appears short in all patients, and a high nasal bridge is present in patients 1, 2, 3, and 4. The ears are generally large but well developed. The additional features found in at least two individuals were chest-wall deformity (pectus carinatum in one patient and pectus excavatum in another) and long and tapering fingers in two individuals, with fifth-finger clinodactyly in one of them. One additional child was noted to have fifth-finger brachydactyly, but X-ray assessment was not performed. An ungainly, ataxic gait was described in two patients, and, in patient 2, diagnoses of Angelman and Rett syndromes were considered and excluded. Each of the remaining features was identified only in a single patient: recurrent bilateral middle ear infections, requiring grommets, adenoidectomy, and removal of a cholesteatoma, in patient 2; ligamentous laxity in patient 3; downslanting palpebral fissures, posteriorly rotated ears, smooth philtrum, and prominent lower lip in patient 4; left-sided unilateral cleft lip and palate and nail hypoplasia in patient 5; progressive microcephaly and abnormal skin pigmentation in patient 6; and horseshoe kidney and hypospadias in the patient described by Rossi et al. (Rossi et al., 2001Rossi E Piccini F Zollino M Neri G Caselli D Tenconi R Castellan C Carrozzo R Danesino C Zuffardi O Ragusa A Castiglia L Galesi O Greco D Romano C Pierluigi M Perfumo C Di Rocco M Faravelli F Dagna Bricarelli F Bonaglia M Bedeschi M Borgatti R Cryptic telomeric rearrangements in subjects with mental retardation associated with dysmorphism and congenital malformations.J Med Genet. 2001; 38: 417-420Crossref PubMed Google Scholar).Table 1Summary of the Clinical Features Found in the Six New Patients with the 3q29 Deletion and in the Patient from the Study by Rossi et al. (Rossi et al., 2001Rossi E Piccini F Zollino M Neri G Caselli D Tenconi R Castellan C Carrozzo R Danesino C Zuffardi O Ragusa A Castiglia L Galesi O Greco D Romano C Pierluigi M Perfumo C Di Rocco M Faravelli F Dagna Bricarelli F Bonaglia M Bedeschi M Borgatti R Cryptic telomeric rearrangements in subjects with mental retardation associated with dysmorphism and congenital malformations.J Med Genet. 2001; 38: 417-420Crossref PubMed Google Scholar)Clinical FeaturePatient 1Patient 2Patient 3Patient 4Patient 5Patient 6Patient from Rossi et al. (Rossi et al., 2001Rossi E Piccini F Zollino M Neri G Caselli D Tenconi R Castellan C Carrozzo R Danesino C Zuffardi O Ragusa A Castiglia L Galesi O Greco D Romano C Pierluigi M Perfumo C Di Rocco M Faravelli F Dagna Bricarelli F Bonaglia M Bedeschi M Borgatti R Cryptic telomeric rearrangements in subjects with mental retardation associated with dysmorphism and congenital malformations.J Med Genet. 2001; 38: 417-420Crossref PubMed Google Scholar)PregnancyNormalNormalNormalNormalBirth weight (kg)3.13.1202.9302.7782.3Length of gestation38 wkTermTerm41 wk41 wkFeedingPoor sucklingFailure to thriveAge at sitting9 mo7 moAge at walking18 mo16 moNormal21 moBy 3 yearsAge at first words28 mo19 moAge when developmental delay was noted18 mo3 years18 mo3 years5 moAge at talking4–5 years6–7 years3 years2 words and Makaton at 8 years, 10 moType of schoolingSpecial schoolSpecial schoolSpecial schoolSpecial school (IQ = 70)Learning support in mainstream schoolSpecial schoolModerate mental retardationHead circumference (percentile)9th–25th3rd50th10thProgressive microcephalyaHead circumference at birth: 32 cm (2 SD below the mean); at age 0.4 years: 37 cm (>4 SD below the mean); and at age 8.8 years: 46 cm (>5 SD below the mean).Height (percentile)25th25th–50th25th40th3rd–10th.4thWeight (percentile)2nd–9th25th3rd–10th.4thHands and feetLong and tapering fingersFifth-finger brachydactylyLong and tapering fingers; clinodactyly of fifth finger and third, fourth, and fifth toesLongBehaviorICD-10 autisticAutistic featuresPectusExcavatumCarinatumGaitArm flapping when youngAtaxia and excited, stereotypical wavingUngainly gait but neurologically normalAdditional featuresChronic otitis media and cholesteatomaLigamentous laxityScaphoid skull (familial) and frontal bossing; downslanting palpebral fissures, smooth philtrum, and prominent lower lip; posteriorly rotated earsProminent metopic suture; left unilateral cleft lip and palate; nail hypoplasiaIncreased pigmentation on dorsum of left foot; not toilet trained; frequent headachesHorseshoe kidney and hypospadiasa Head circumference at birth: 32 cm (2 SD below the mean); at age 0.4 years: 37 cm (>4 SD below the mean); and at age 8.8 years: 46 cm (>5 SD below the mean). Open table in a new tab Molecular cytogenetic results are summarized in table 2 and figure 2. The samples were collected from laboratories in the United Kingdom and the Netherlands (Nijmegen), and FISH was performed on interphase or metaphase spreads, depending on the sample availability. G-banding at the 550 band or at a higher level, FISH probe preparation, and hybridization were performed using standard techniques. Parental samples were available for patients 1–5, and all deletions had arisen de novo. Patient 6 was adopted out of the birth family; thus, parental samples were unavailable for analysis. FISH analysis of patients 1 and 2 showed the presence of a terminal deletion of 3q29 by use of 3qter probe pVYS223B from the Totelvysion subtelomere screening kit (Vysis) on metaphase spreads. Deletions in patients 3 and 6 were initially suspected on high-resolution G-banding and were confirmed using the 3q subtelomeric probe CTC-196F4 from the study by the National Institutes of Health and Institute of Molecular Medicine Collaboration (National Institutes of Health and Institute of Molecular Medicine Collaboration, 1996National Institutes of Health and Institute of Molecular Medicine Collaboration A complete set of human telomeric probes and their clinical application.Nat Genet. 1996; 14: 86-89Crossref PubMed Scopus (291) Google Scholar). In patient 5, the deletion was detected using probe 3qtel106 from the Cytocell subtelomere screening kit. Probes pVYS223B, CTC-196F4, and 3qtel106 all map within BAC clone RP5-1061C18 (fig. 2A). In patient 4, the deletion was detected using multiplex ligation-dependent probe amplification (MLPA) (Koolen et al. Koolen et al., 2004Koolen DA Nillesen WM Versteeg MH Merkx GF Knoers NV Kets M Vermeer S van Ravenswaaij CM de Kovel CG Brunner HG Smeets D de Vries BB Sistermans EA Screening for subtelomeric rearrangements in 210 patients with unexplained mental retardation using multiplex ligation dependent probe amplification (MLPA).J Med Genet. 2004; 41: 892-899Crossref PubMed Scopus (135) Google Scholar). The 3q29 deletion was not detectable with the older MLPA MRC (Medical Research Council) Holland telomere kit, PO19, which recognizes sequence in the nondeleted clone RP11-496H1, but is identifiable only with the new kit, PO36, which uses sequence within the BDH gene located within the deleted clones RP13-616I3 and RP11-535N19 as a MLPA probe. To refine the deletion, 26 overlapping BAC and PAC clones were identified using the then-available build 31 sequence (National Center for Biotechnology Information [NCBI] Map Viewer). The human genomic clones from the RPCI-5, -11, and -13 BAC and PAC libraries were obtained from the BACPAC Resources Center. Slides were analyzed using a fluorescence microscope (Leica DMRB), and images were recorded using SmartCapture 2 software (Digital Scientific). Published STS primers STS-R44803, SHGC-170324, D3S4248, SHGC-34823, SHGC-149375, SHGC-10638, RH45269, SHGC-146015, RH99161, D3S2320, and RH80465 were used to check the identity of the clones (fig. 2B). In patients 1–4 and patient 6, the deletion endpoints were refined to within a single BAC clone. The telomeric breakpoint was within BAC clones RP11-594G13 and RP11-496H1, and the centromeric breakpoint was within BAC clones RP11-185G19 and RP11-480A16. For patient 5, there was insufficient material available to define the centromeric breakpoint, but the telomeric breakpoint was within BAC clone RP11-496H1, as for patient 2. The centromeric breakpoint lay between BAC clones RP11-171N2 and RP11-185G19 (table 2). At the telomeric breakpoint, the two BAC clones that define the deletion endpoint in the six patients overlap, and both contain STSs RH45269 and SHGC-146015. Most of clone RP11-594G13 is also contained within RP11-496H1, but clone RP11-594G13 extends ∼60 kb further toward the centromere than RP11-496H1. On the basis of the resolution of FISH, the breakpoints in each of the patients are likely to be similar and to lie within 60–70 kb of each other.Table 2Limits of Interstitial 3q29 Microdeletion in Six Unrelated PatientsPresence or Absenceof Probe in PatientbP = probe was present on both chromosome 3 homologues by FISH; D = probe was deleted on one of the chromosome 3 homologues; NT = probe was not tested, because of insufficient chromosomal material available from the patient.Contig andBAC/PAC CloneaUnless otherwise stated, the BAC or PAC clones used were from the RP11 library.AccessionNumber123456NT_005612: 513G11AC117469PPPPPPNT_005535: 279P10AC125362PPPPPPNT_029928: 171N2AC069513PPPPNTP 352G9AC124944PPPPNTP 480A16AC024937PDDDNTD 185G19AC139666DDDDNTD 252K11AC026308DDDDDD 447L10AC069257DDDDNTD 106N22AC083822DDDDNTD 200I19AC092933DDDDNTD 133B21AC023797DDDDNTD 470E12AC055725DDDDNTD 607N15AC127904DDDDNTD 778E2AC016949DDDDDD 432D10AC068302DDDDNTD 114F20AC092937DDDDDD RP5-1061C18AL121981DDDDDD RP13-616I3AC128709DDDDDD 535N19AC126183DDDDDD 594G13AC132008PDPPDP 496H1AC024560PPPPPP 803P9AC055764PPPPPP 23M2AC022621PPPPPP 237O3AC144530PPPPNTP 643E20AC135893PPPPNTP 694O4AC073135PPPPNTPNote.—All accession numbers listed in the table are from GenBank.a Unless otherwise stated, the BAC or PAC clones used were from the RP11 library.b P = probe was present on both chromosome 3 homologues by FISH; D = probe was deleted on one of the chromosome 3 homologues; NT = probe was not tested, because of insufficient chromosomal material available from the patient. Open table in a new tab Note.— All accession numbers listed in the table are from GenBank. The extent of the common microdeletion was determined using the finished sequence clones selected from the contigs on NCBI build 35.1 (NCBI Map Viewer). The microdeletion is estimated to be ∼1.5 Mb, and it contains at least 22 transcripts, 5 of which are known genes (PYT1A, PAK2 [MIM 605022], MFI2 [MIM 155750], DLG1 [MIM 601014], and BDH [MIM 603063]), 7 of which are incomplete cDNAs with two or more documented cDNA sequences, and 10 of which are hypothetical genes with no experimental evidence. Since the deletion limits in the six patients were almost identical, the finished genomic sequence on each side of the deletion was investigated to determine whether any region-specific low-copy repeats (LCRs) were present. Nix analysis (HGMP-RC Nix Session Web site) of RP11-496H1 (GenBank accession number AC024560) found regions of sequence homology within BAC clones RP11-480A16, RP11-352G9, and RP11-171N2. Two separate LCR sequences, designated “repeat A” and “repeat B,” were identified on each side of the deletion region (fig. 2C). By use of BLAT analysis (Human BLAT Search Web site), repeat A was identified four times on chromosome 3, at positions 198832974–198852444, 197195390–197215144, and 197150598–197155588 in one orientation and at position 196868577–196884133 in the opposite orientation. Each repeat was >97.5% homologous and was ∼19 kb, ∼19 kb, ∼5 kb, and ∼15 kb long, respectively (fig. 2C). Repeat B occurred twice on chromosome 3; both repeats were in the same orientation and were located on chromosome 3 on each side of the breakpoint, at positions 198860847–198872158 and 197160541–197171848. These sequences were 98% homologous, and both were 11 kb in length (fig. 2C). In summary, there is evidence of the presence of region-specific LCRs within the genomic sequence at each end of the microdeletion. It is likely that the formation of the similarly sized de novo microdeletions identified in the six patients was facilitated by these repeats and probably arose by nonallelic homologous recombination between LCRs on each side of the breakpoint, resulting in a single copy of the LCR and a genomic deletion between them. Alternatively, the presence of LCRs on each side of the deletion may not be directly involved in the deletion, per se, but they may act to predispose the genome to form a deletion. Both disease-causing mechanisms are now well recognized, although the presence of LCRs on each side of the deletion suggests that the former mechanism is more likely (Osborne et al. Osborne et al., 2001Osborne LR Li M Pober B Chitayat D Bodurtha J Mandel A Costa T Grebe T Cox S Tsui LC Scherer SW A 1.5 million-base pair inversion polymorphism in families with Williams-Beuren syndrome.Nat Genet. 2001; 29: 321-325Crossref PubMed Scopus (245) Google Scholar; Shaw and Lupski Shaw and Lupski, 2004Shaw CJ Lupski JR Implications of human genome architecture for rearrangement-based disorders: the genomic basis of disease.Hum Mol Genet. 2004; 13: R57-R64Crossref PubMed Google Scholar). This analysis of six patients with an interstitial microdeletion of 3q29 is the first collation of cases to delineate 3q29 microdeletion syndrome. The clinical features of a patient from a previous report have been included for comparison, but samples from this patient were not available to establish the exact deletion breakpoints, and the family has been lost to clinical follow-up (E. Rossi, personal communication). The molecular deletion boundaries detected in the patients are strikingly similar, yet there is considerable clinical variability in the phenotype, with mild-to-moderate mental retardation being the only common and consistent feature. In addition, slightly dysmorphic facial features are similar in most patients: a long and narrow face, short philtrum, and high nasal bridge. As additional patients are screened for microdeletion syndromes on the basis of mental retardation and autism alone, it is likely that this syndrome will become increasingly well identified. Understanding the exact molecular mechanism of disease in these patients—that is, understanding how the deletion of some 22 genes in this region affects the development of an individual—is, of course, the next challenge. At this stage, it is impossible to attribute the phenotype to any one of the deleted genes, but two genes within the deleted area—PAK2 and DLG1—merit further interest, since both are autosomal homologues of known X-linked mental retardation genes PAK3 (MIM 300142) and DLG3 (MIM 300189). Loss-of-function mutations in either PAK3 or DLG3 result in moderate-to-severe mental retardation (Allen et al. Allen et al., 1998Allen KM Gleeson JG Bagrodia S Partington MW MacMillan JC Cerione RA Mulley JC Walsh CA PAK3 mutation in nonsyndromic X-linked mental retardation.Nat Genet. 1998; 20: 25-30Crossref PubMed Scopus (380) Google Scholar; Tarpey et al. Tarpey et al., 2004Tarpey P Parnau J Blow M Woffendin H Bignell G Cox C Cox J et al.Mutations in the DLG3 gene cause nonsyndromic X-linked mental retardation.Am J Hum Genet. 2004; 75: 318-324Abstract Full Text Full Text PDF PubMed Scopus (136) Google Scholar). The DLG1 protein SAP97, like SAP102 (DLG3), is a component of the postsynaptic density, and RNAi knockdown experiments of SAP97 result in reduced surface expression of GRIA1 (MIM 138248) and GRIA2 (MIM 138247) and a decrease in both AMPA (MIM 138248) and NMDA (MIM 138251) excitatory postsynaptic currents (Nakagawa et al. Nakagawa et al., 2004Nakagawa T Futai K Lashuel HA Lo I Okamoto K Walz T Hayashi Y Sheng M Quaternary structure, protein dynamics, and synaptic function of SAP97 controlled by L27 domain interactions.Neuron. 2004; 44: 453-467Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar). This suggests that loss of PAK2 or DLG1 may have a critical role in the development of mental retardation in these patients, but, clearly, this hypothesis needs further investigation. We are grateful to the patients and families, for taking part in this study, and to the clinicians, for referring them. We thank Elizabeth Kerr for her expert technical assistance. J.C. is supported by the Medical Research Council, United Kingdom; H.M. is supported by the Department of Health, United Kingdom; B.B.A.d.V. is supported by a grant from the Netherlands Organization for Health Research and Development; and F.L.R. is supported by the Wellcome Trust.