Figure 1.
All FOXL2 encompassing deletions were initially identified using MLPA. The regulatory deletions were identified using a combined approach of microsatellite analysis and a custom-made quantitative PCR assay of the FOXL2 region (qPCR-3q23) [42]–[44]. For further delineation of the deletions two different array-based methods were used in a first step: (1) custom high-resolution 8×60 K Agilent microarrays for 35 deletions at the CMGG, and (2) genome-wide Illumina Human610-Quad BeadChip arrays for 7 deletions at the INGEMM. Subsequently, long-range PCR was performed if the sum of the breakpoint regions was smaller than 15 kb. However, if the sum of breakpoints was larger than 15 kb, the breakpoint regions were first further delineated using a qPCR-based copy number screening approach. Long-range PCRs resulting in a specific junction product underwent sequencing with internal primers. Finally, several bioinformatic tools were used in order to determine the underlying deletion mechanism.
Figure 2.
Overview of the delineated regulatory and FOXL2 encompassing deletions.
Overview of the FOXL2 region (chr3:135099979–142458004; UCSC, Human Genome Browser, hg19) with custom tracks showing the delineated regulatory and FOXL2 encompassing deletions presented in this study, numbered from A to H and from 1 to 16 respectively. The horizontal red bars indicate the deleted regions (regulatory deletions are shown in dark red and FOXL2 encompassing deletions are shown in light red). At the top, the RefSeq Genes track is included. The locations of FOXL2 and long non-coding RNA PISRT1 are indicated by vertical blue and yellow lines respectively. Additional information on genes (i) contained in the deletion, (ii) spanning the breakpoints, or (iii) located outside the respective deletion and their distances to the breakpoint, can be found in Table S1.
Figure 3.
Schematic representations of the genomic architecture for 5 exemplary regulatory and FOXL2-encompassing deletion.
For deletions A, 1, 5, 7 and 12, both breakpoint regions joined by the deletion are shown. These deletions were selected as an example for each group (group 1: deletion A, group 2: deletion 7, and group 3: deletions 1, 5 and 12) of most likely molecular mechanism as described in the discussion. A breakpoint region is displayed as the combination of two colored, solid lines together representing a 150 bp DNA sequence. The proximal breakpoint region consists of a non-deleted blue line and a deleted red line while the distal breakpoint region consists of a deleted red line and a non-deleted green line. Each deletion is composed of the two red, solid lines joined by the red dashed line which represents the different size of the deletion for every patient. The actual size of the deletions is indicated above the red, dotted lines. The pink vertical arrows mark the position of the breakpoints displaying the number of base pairs of microhomology between both breakpoint regions and the junction product (see also Figure 4 and Figure S3). The presence of repetitive elements is shown as bars of different shades of gray (Alu elements are shown in light grey bars, other repetitive elements are shown in dark grey bars). Sequence motifs are indicated with orange, skewed lines intersecting with the sequence. Direct repeats, oligo(G)n tracts and Z-DNA are represented by dark purple arrows, dark purple bars and light purple bars respectively. The schematic representations for the other deletions can be found in the online supplement (Figure S2).
Table 1.
Overview of bioinformatic results.
Figure 4.
Multiple sequence alignment of 4 exemplary junctions.
The junctions of deletion A (A), 4 (B), B (C) and 11 (D) are shown as an example for the different lengths of microhomology. Sequences of 150 bp surrounding each junction are aligned to the proximal and distal reference sequences using ClustalW. The proximal and distal reference sequences are shown in blue and green respectively. The junction sequences are depicted in the colour of the reference sequence they align with. Microhomology between the proximal and distal reference sequence and the junction are shown in pink. The other multiple sequence alignments can be found in the online supplement (Figure S3).
Figure 5.
Distribution pattern of microhomology.
Bar chart displaying the distribution of microhomology in the random control population (purple) and the observed breakpoints in this study (blue). Microhomology in the random control population clusters around 0 to 1 bp, while longer stretches of microhomology are noted for the observed breakpoints.