Single-molecule analysis of genome rearrangements in cancer

Abstract

Rearrangements of the genome can be detected by microarray methods and massively parallel sequencing, which identify copy-number alterations and breakpoint junctions, but these techniques are poorly suited to reconstructing the long-range organization of rearranged chromosomes, for example, to distinguish between translocations and insertions. The single-DNA-molecule technique HAPPY mapping is a method for mapping normal genomes that should be able to analyse genome rearrangements, i.e. deviations from a known genome map, to assemble rearrangements into a long-range map. We applied HAPPY mapping to cancer cell lines to show that it could identify rearrangement of genomic segments, even in the presence of normal copies of the genome. We could distinguish a simple interstitial deletion from a copy-number loss at an inversion junction, and detect a known translocation. We could determine whether junctions detected by sequencing were on the same chromosome, by measuring their linkage to each other, and hence map the rearrangement. Finally, we mapped an uncharacterized reciprocal translocation in the T-47D breast cancer cell line to about 2 kb and hence cloned the translocation junctions. We conclude that HAPPY mapping is a versatile tool for determining the structure of rearrangements in the human genome.

Figure 1.

Linkage information is needed to determine the structure of some genome rearrangements. (A–C) Two rearrangements that cannot be distinguished by array-CGH. (A) Array-CGH profile that would be obtained for the black chromosome from either (B) or (C). (B) A small deletion in the black chromosome; (C) a reciprocal translocation has resulted in net loss of a small section of the black chromosome at the translocation breakpoints. (D–G) Two rearrangements that cannot be distinguished either by array-CGH or by finding breakpoint junctions by large-scale sequencing (5,6), but could be distinguished by HAPPY mapping, provided the breakpoints are within 1 Mb of each other. (D) Array-CGH data for the black chromosome obtained from either (F) or (G); (E) junction sequences obtained from either (F) or (G); (F) a piece of the black chromosome is inserted between pieces of the grey and white chromosomes; while in (G), there are two separate translocations of the black chromosome with respectively the grey and white chromosomes, both including the region between the dotted lines. (The grey and white ‘chromosomes’ could also represent other parts of the black chromosome, as in an inversion, for example).

Figure 2.

Combined HAPPY mapping and molecular copy-number counting gives positional information and copy number. (A) The circles represent marker sequences on chromosomes in a normal cell. (B) An unbalanced translocation leaves one copy of each normal chromosome and one copy of a hybrid chromosome (grey). (C) DNA is prepared and broken at random; each fragment is large enough to span several markers. (D) Fragmented DNA is dispensed into samples (typically 88), each containing about half a genome’s worth of fragments. (E) Each sample is scored, usually by multiplexed PCR, for the presence (ticks) of each marker. The red and yellow markers always occur together (co-segregate), since they remain adjacent on all chromosomes; the red and green markers cosegregate often (four of the samples) but not always (four samples contain the red marker without the green), since they remain adjacent on the normal chromosome but not the hybrid chromosome; likewise, the red and black markers, which are brought together on the hybrid chromosome, occur together in about half of the samples. Other pairs of markers that are never adjacent (e.g. black and green; blue and red) occur together only occasionally, by chance. (F) The number of samples that score positive for any given marker also reflects the relative number of copies in the genome; in this idealized example, the green and blue markers are present at only half the copy-number of the others. This way of measuring copy number has been described previously as molecular copy-number counting (12).

Figure 3.

Distinguishing between simple deletion and loss at a rearrangement junction. (A, B) HAPPY mapping indicates that copy number loss at 36.5 Mb on chromosome 8 in T-47D is an interstitial deletion; (A) part of raw data. Rows are PCR markers, vertical scale is genome position. Columns are 20 (of 88) diluted DNA samples. Horizontal lines are PCR markers and the individual dashes represent PCR results; positive hits are joined by vertical black bars, representing the presence of DNA. The possible deletion is shaded. Evidence that this is a simple interstitial deletion comes from the concordance of markers that flank the copy number loss, i.e. markers are positive on both sides or negative on both sides. This is expressed (B) as linkage between markers, calculated as the log of odds (LOD) that the markers are linked, using all 88 samples. Horizontal lines represent markers, loops (arcs) represent linkage between them: the stronger the linkage, the further the loop extends to the right. For clarity, only linkage LOD > 7 is shown, and linkage to the markers within the deletion is omitted. (C) Copy-number loss at 38.2 Mb on chromosome 8. Linkage LOD > 5 is shown. In contrast to (A, B), this is loss at an inversion junction, and there is no linkage of this strength across the copy number loss, even though, because DNA fragments were large, linkage over flanking sequences extends >100 kb.

Figure 4.

Complex translocation of chromosomes 8 and 14 in T-47D with inversion and losses of chromosome 8. This was characterized previously, by fluorescence in situ hybridization (FISH) and cloning of junction sequences (13). (A) Normal chromosomes 14 (long arm only) and 8 (short arm only). Dotted lines, breakpoints of translocation; small circles, key points on the chromosomes; large circles, centromeres. (B) Rearranged chromosome. The distal part of 14 has been translocated to 8, but, in addition, 3.5 Mb of chromosome 8 adjacent to the translocation has been inverted, and about 100 kb has been lost at the inversion junction (indicated by the gap and open arrowhead). The black arrowhead marks the further copy-number loss at 36.5 Mb, investigated in Figure 3A, which may be on this chromosome or on the normal chromosome 8. (C) Resulting copy-number profile, showing the two copy-number losses. There are two copies of the 8;14 translocation, two normal chromosome 8s and two normal 14 s, per cell, as T-47D is pseudo-tetraploid (13,16). [Adapted from ref. (13)].

Figure 5.

Detection of changed linkage at a chromosome translocation. (A) Schematic of translocation between chromosome 1 and 8, previously mapped using array painting (17), in breast cancer cell line HCC1187. On the left, normal chromosomes 1 and 8 with points of breakage A, B and C (dashed horizontal lines). Right, resulting products and piece of chromosome 1 that is lost. Dotted vertical lines show (not to scale) markers used for mapping. There are two copies per cell of both translocated chromosomes, and additional untranslocated copies of the breakpoint regions. In consequence, there are three copies throughout the chromosome 8 region, but a transition from four to two copies at the chromosome 1 breakpoint. (B) HAPPY mapping around the breakpoints. The heavy horizontal lines represent the reference genome to scale; markers are named below. The arcs indicate linkage between markers as LOD scores as before; linkages between markers normally on the same chromosome are above the line; linkages between markers on different chromosomes are below. LOD values >2 are shown. The arrows indicate the largest losses of linkage compared to that expected for the normal genome, which are likely to indicate the breakpoints. Copy-number results are shown below the marker names as vertical lines indicating relative copy number, expressed as mean number of copies per sample of the mapping panel; the horizontal lines joining these show the average values across the markers they cover.

Figure 6.

Establishing linkage between rearrangement junctions found by paired-end sequencing. (A) Three junctions, J1–J3, which were identified by paired-end sequencing or cloning, that join points on chromosomes 11 and 16. Blue and magenta bars, regions of normal chromosome 11 and 16 showing where the breakpoints of the junctions map. Arcs show the junctions, numbers are genomic positions of breakpoints in basepair. (B) Four possible ways in which these junctions could be joined together. The blue and magenta bars correspond to copies of the regions of chromosomes 11 and 16 shown in (A). Coloured-in parts of the bars are present in the postulated rearranged chromosome, whereas regions that are not present are not coloured in. For example, in (B)(i) (which is the correct model), in the upper part, chr. 11 is broken at 10.4 Mb and the fragment extending from the break towards the q telomere is joined at J3 to a piece of chr. 16 extending from a break at 66.14 Mb towards the q telomere; while on a separate chromosome, chr.11 up to a break at 9.1 Mb is joined to the 1.4-kb fragment of chr. 11 between J1 and J2 (10.5195–10.5209 Mb), and then joined to a piece of chr. 16 extending up to the breakpoint at 66.19 Mb. (C) Relationship between linkage and genomic distance, established for markers on a control region of the genome. Note that at around 50–60 kb separation, LOD scores between 1.9 and 7 were obtained between various combinations of control markers. (D) LOD scores obtained between the junction-specific markers. Linkage between J1 and J2 had a LOD of 31, i.e. these junctions are very close, whereas no significant linkage was found between J2 and J3, which should have shown a LOD score between 1.9 and 7 if they were 55 kb apart on the same chromosome. These scores support the arrangement shown in B(i), as expected.

Figure 7.

Mapping a reciprocal chromosome translocation t(10;20)(q21;q13.2) in the T-47D cell line. (A) Diagram of translocation. (B) HAPPY mapping shows linkage of markers on chromosome 10 (left side of diagram) to markers on chromosome 20 (right side), and loss of linkage across the breakpoints on each chromosome (dotted red lines). As in Figure 5, the heavy blue horizontal lines represent the reference genome to scale; the arcs indicate linkage between markers as LOD scores as before; linkages between markers normally on the same chromosome are below the line; linkages between markers on different chromosomes are above. Linkage arcs in green join the two parts of the derivative 10 chromosome, der(10)t(10;20)(q21;q13.2) (i.e. the product chromosome with the chromosome 10 centromere); linkage arcs in red join the parts of the reciprocal product der(20)t(10;20)(q21;q13.2). LOD values >2 are shown. Marker names include their chromosome and position of Fex primer in bp. Above the marker names are vertical lines indicating relative copy number. Insets (C) and (D): the breakpoint regions with genomic scale stretched 6-fold. Data shown was from a final mapping run in which all markers were used together, for illustration purposes.