Evolutionary history of mammalian transposons determined by genome-wide defragmentation

Abstract

The constant bombardment of mammalian genomes by transposable elements (TEs) has resulted in TEs comprising at least 45% of the human genome. Because of their great age and abundance, TEs are important in comparative phylogenomics. However, estimates of TE age were previously based on divergence from derived consensus sequences or phylogenetic analysis, which can be unreliable, especially for older more diverged elements. Therefore, a novel genome-wide analysis of TE organization and fragmentation was performed to estimate TE age independently of sequence composition and divergence or the assumption of a constant molecular clock. Analysis of TEs in the human genome revealed approximately 600,000 examples where TEs have transposed into and fragmented other TEs, covering >40% of all TEs or approximately 542 Mbp of genomic sequence. The relative age of these TEs over evolutionary time is implicit in their organization, because newer TEs have necessarily transposed into older TEs that were already present. A matrix of the number of times that each TE has transposed into every other TE was constructed, and a novel objective function was developed that derived the chronological order and relative ages of human TEs spanning >100 million years. This method has been used to infer the relative ages across all four major TE classes, including the oldest, most diverged elements. Analysis of DNA transposons over the history of the human genome has revealed the early activity of some MER2 transposons, and the relatively recent activity of MER1 transposons during primate lineages. The TEs from six additional mammalian genomes were defragmented and analyzed. Pairwise comparison of the independent chronological orders of TEs in these mammalian genomes revealed species phylogeny, the fact that transposons shared between genomes are older than species-specific transposons, and a subset of TEs that were potentially active during periods of speciation.

Figure 1. Defragmentation of an ∼11 kb TE Cluster by TCF

A window from the UCSC genome browser showing an 11-kb transposon cluster from Chromosome 20p12.3. The uploaded custom track output provided by TCF and the output from the Repeat Masker track are shown. An L2 LINE (on the minus strand) has been interrupted by two TEs, a DNA transposon MER63, and a LINE L1MB3. The MER63 element has in turn been interrupted by an AluY element. The L1MB3 has been interrupted by a DNA transposon Tigger1 element, which has in turn been interrupted by a LINE L1PA8, which in turn has been interrupted by two Alu Y elements. This analysis reveals the evolutionary history of this region of the genome by defragmentation of TE clusters. The cluster table (bottom) shows Repeat Masker data for each TE fragment collected by TCF. Columns in the table are Genome Start, Genome End (starting and ending hg18 genomic coordinates for each TE fragment), Strand, % Divergence (of TE fragment from consensus), Repeat Start, Repeat End, Repeat Left (repeat indices relative to the derived consensus), and Name, Family, and Class (of each TE fragment).

Figure 2. Interruption Matrix Analysis of the Chronological Order of TEs

(A) A 360 × 360 adjacency matrix showing the number of times that each of the 360 human TEs interrupt each of the other 360 TEs, with values represented as a heat map as indicated. The TEs are shown in the same order on both the horizontal and vertical axes. This matrix shows a random TE order, and has an upper triangle matrix penalty score of ∼45,000. (B) Schematic of the matrix with TEs arranged in the correct chronological order from oldest to youngest (decreasing age) on both the horizontal axis (left to right) and the vertical axis (top to bottom). The corners of this matrix will contain points that represent old into old TEs (top left), new into old TEs (bottom left), and new into new TEs (bottom right). New into old TEs (top right) are not expected. This forms the basis for the objective function, which minimizes the upper triangle matrix by element repositioning (see text). (C) The 360 × 360 adjacency matrix after performing the repositioning algorithm. This represents one solution from one starting order, with a penalty of ∼7,800. There are 360! possible orders, which represents a state space that is far too large (∼10⁵⁰⁰ orders) to search completely. (D) Graphical illustration of the results for three TEs of different relative ages. For each TE, the pink graph shows the amount that the TE has interrupted the other elements (interrupTER row in the matrix), and the blue graph shows the amount that the TE has been interrupted by other TEs (interrupTEE column in matrix). The TEs are arranged along the horizontal axis in the final chronological order as determined by IMA. The MLT1J element (top) is relatively old (position 32), and interrupts only a few relatively old elements (pink), but is interrupted by many newer elements (blue). The MLT1B element (middle) is of intermediate age (position 154), and gets interrupted by newer elements (blue) and interrupts older elements (pink) in similar amounts. The AluSx (bottom) is relatively new (position 317), and interrupts many older elements (pink) but is only interrupted by a few newer elements (blue). The values in these graphs have been normalized as described in Materials and Methods. A polynomial trend line of power 3 (black curve) is fitted to each set of points.

Figure 3. Chronological Ordering of Human TEs

(A) Chronological order of 360 human TEs as derived by IMA. The individual names of the elements are not visible in Figure 3 (see Table S2 for full dataset). The range of positions of several TE families is shown (yellow bars) to illustrate the agreement with previous phylogenetic age analyses. The names and positions of the Charlie and Tigger families of DNA transposons are shown. (B) Position of individual elements from the L1 subfamilies L1ME, L1MB, and L1PA. Also shown is the median percent divergence from the Repeat Masker–derived consensus sequence (Table S2), and the percent connectedness of this TE in the matrix (see text and Table S2). In general, the percent divergence agrees well with the relative age of the element as determined by IMA. (C) The positional distribution for the TEs listed in (B) is displayed on the horizontal axis, relative to the overall chronological order of 360 elements on the vertical axis. (Thin line, range of the lowest 5% or highest 5% of positions calculated; thicker line, range of the next lowest 20% or next highest 20% of positions calculated; thickest line, range of middle 50% of positions calculated; black bar, median position). The chronological order in (A) was derived by ordering the medians of the positional distributions.

Figure 4. Analysis of DNA Transposon Activity in the Human Genome

The chronological order for MER1 (red) and MER2 (green) families of DNA transposons, taken from the 405 elements used in this run of IMA, is shown on the left (vertical axis) from oldest (top) to youngest (bottom). The positional distribution is shown on the horizontal axis. The name of each element, its position in the chronological order (out of 405 elements), the median percent divergence (Table S2), and the percent connectedness (Table S2) are shown. The two major radiations of MER1 and MER2 DNA transposons can be seen. Subfamilies of DNA transposons containing autonomous parental elements and dependent nonautonomous elements are indicated. The following DNA transposons were not included in this figure because of either low connectedness and/or long positional distribution, for space and clarity (position name; % divergence; connectedness): (37 MER102a- 27.9; 5), (41 MER69b- 25.1; 20), (45 MER91- 23.9; 6), (59 MER91b- 25.8; 9), (72 MER97c- 23.6; 33), (85 MER117- 28.1; 15), (95. MER69a- 24.3; 10), (103 MER91a- 30.4; 12), (105, MER91c- 26.5; 9), (106 MER58d- 19.9; 10), (110 MER97a- 22.6; 7), (117 MER97b- 24.4; 5), (143 Tigger6b- 19.0; 5), (127 MER4–5 21.4; 15), (125 MER45-r 21.0; 13), and (210 pMER- 15.9; 0).

Figure 5. Comparison of Transposon History in Seven Mammalian Genomes

A colorimetric pairwise comparison of the chronological order of the TEs from seven mammalian genomes. Each of the seven species is indicated by a different color. Each species is shown in turn as the reference genome (indicated on top), with the chronological order derived from IMA shown on the left. For each reference genome, the TEs from the other six genomes are shown aligned to the reference genome. TE names in reference genomes have been replaced in this figure by solid-color bars due to space considerations (see Figure S2 for TE names and full details). When a TE in the aligned genome matches the position in the chronological order of the reference genome (based on the criteria described in Materials and Methods), it is indicated by a solid-color bar corresponding to the species, and when it does not match the position (but is found in the reference genome), it is indicated by a lighter (stippled) color bar. A gray bar indicates that the TE was in the reference genome, but is not found at all in the aligned genome. A black bar indicates the TE was in both the reference and aligned genome, but was not ordered in the aligned genome (not in the set used for IMA analysis; see Materials and Methods). The total count of the matching (solid), not matching (stippled), not in genome (gray), and not in set (black) bars are shown in Table S4 for each reference genome. Subsequent to this pairwise comparison, the list of TEs from each genome were aligned to the human genome by insertion of spaces (white bars), which helps to maintain the position of TEs down the lists in order to allow comparison across the different species. After the point of divergence of the mouse and rat clade from the rest of the mammalian genomes, the rat genome is aligned to the mouse. Neighbor-joining tree of the seven mammals is shown at the bottom. Note that the tree is unrooted.