Dosage-dependent severity of the phenotype in patients with mental retardation due to a recurrent copy-number gain at Xq28 mediated by an unusual recombination

Abstract

We report on the identification of a 0.3 Mb inherited recurrent but variable copy-number gain at Xq28 in affected males of four unrelated families with X-linked mental retardation (MR). All aberrations segregate with the disease in the families, and the carrier mothers show nonrandom X chromosome inactivation. Tiling Xq28-region-specific oligo array revealed that all aberrations start at the beginning of the low copy repeat LCR-K1, at position 153.20 Mb, and end just distal to LCR-L2, at 153.54 Mb. The copy-number gain always includes 18 annotated genes, of which RPL10, ATP6AP1 and GDI1 are highly expressed in brain. From these, GDI1 is the most likely candidate gene. Its copy number correlates with the severity of clinical features, because it is duplicated in one family with nonsyndromic moderate MR, is triplicated in males from two families with mild MR and additional features, and is present in five copies in a fourth family with a severe syndromic form of MR. Moreover, expression analysis revealed copy-number-dependent increased mRNA levels in affected patients compared to control individuals. Interestingly, analysis of the breakpoint regions suggests a recombination mechanism that involves two adjacent but different sets of low copy repeats. Taken together, our data strongly suggest that an increased expression of GDI1 results in impaired cognition in a dosage-dependent manner. Moreover, these data also imply that a copy-number gain of an individual gene present in the larger genomic aberration that leads to the severe MECP2 duplication syndrome can of itself result in a clinical phenotype as well.

Figure 1

Pedigrees of Families with an Increased Copy Number at Xq28 (A) Family 1, (B) family 2, (C) family 3, and (D) family 4. The result of the qPCR with the GDI1 primer set is indicated below the individual as “N” in case of a normal copy number, and with “+” in case of an increased copy number.

Figure 2

Schematic Representation of the Xq28 Region, 152.90–153.60 Mb, Showing the Recurrent Aberration Detected in Four Unrelated Families (A) Indicated from top to bottom; the location of the qPCR primer pairs (1–17); annotated genes present in the interval; and the LCRs within the region. The amplified region is indicated at the bottom with the horizontal dark bar, which is flanked by the light bars with a normal copy number. The horizontal striped bar represents a copy-number polymorphism identified in a healthy male. (B) Magnification of the start and end positions of the four aberrations. The vertical arrows point to the positions of these sites, demonstrating that they are located at the beginning of LCR K1 and at the end of LCR L2, respectively. The locations of the different regions (R2 to R4, Ki and Li), as referred to in the text, are shown as well.

Figure 3

Custom 44K Oligo-Array-CGH Plots for Analysis of the Copy Numbers within the Region 152.70 Mb to 153.65 Mb at Xq28 (A) Patient IV.2 of family 1. (B) Patient II.1 of family 2. The borders of the copy-number-variable regions (R1 to R4) are indicated by vertical lines with the positions of the breakpoints shown. The mean log2 Cy5/Cy3 ratio (y axis) of each region is given at the right. Below each plot the Ensembl view of the corresponding region is shown with the LCR sets (J, K, L) indicated. At the bottom, we show a schematic representation of the copy number of each region, based on the array-CGH data.

Figure 4

Custom 44K Oligo-Array-CGH Plots for Analysis of the Copy Numbers within the Region 152.70 Mb to 153.65 Mb at Xq28 (A) Patient II.2 of family 3. (B) Patient II.1 of family 4. The borders of the copy-number-variable regions (R1 to R4) are indicated by vertical lines with the positions of the breakpoints shown. The mean log2 Cy5/Cy3 ratio (y axis) of each region is given at the right. Below each plot the Ensembl view of the corresponding region is shown with the LCR sets (J, K, L) indicated. At the bottom, we show a schematic representation of the copy number of each region, based on the array-CGH data.

Figure 5

Recombination Model that Can Explain the Variable Copy-Number Gains at Xq28 (A) The model is based on two consecutive NAHR events in meiosis I. The first is between L1 and L2 on different chromatids directly followed by NAHR between K1 and K2, present on homologous chromatids. (B) In family 2, the copy number was extended to five because of a third NAHR, most likely between sister chromatids in meiosis II.

Figure 6

mRNA Expression Analysis of GDI1, RPL10, ATP6AP1, and FLNA in Affected Members of Families 1 and 2 with a Copy-Number Gain at Xq28 cDNA was prepared from RNA extracted from blood (controls, family 2) or EBV-PBL cell lines (controls, family 1). Compared to controls, all affected individuals showed significant increased mRNA levels for all four genes. Expression was determined by qPCR and normalized to the expression of HPRT. Expression levels are calculated relative to the mean levels obtained in the control samples (fold difference; y axis). Standard deviations of at least two independent experiments are indicated for each bar.