A comprehensive analysis of recently integrated human Ta L1 elements

Abstract

The Ta (transcribed, subset a) subfamily of L1 LINEs (long interspersed elements) is characterized by a 3-bp ACA sequence in the 3' untranslated region and contains approximately 520 members in the human genome. Here, we have extracted 468 Ta L1Hs (L1 human specific) elements from the draft human genomic sequence and screened individual elements using polymerase-chain-reaction (PCR) assays to determine their phylogenetic origin and levels of human genomic diversity. One hundred twenty-four of the elements amenable to complete sequence analysis were full length ( approximately 6 kb) and have apparently escaped any 5' truncation. Forty-four of these full-length elements have two intact open reading frames and may be capable of retrotransposition. Sequence analysis of the Ta L1 elements showed a low level of nucleotide divergence with an estimated age of 1.99 million years, suggesting that expansion of the L1 Ta subfamily occurred after the divergence of humans and African apes. A total of 262 Ta L1 elements were screened with PCR-based assays to determine their phylogenetic origin and the level of human genomic variation associated with each element. All of the Ta L1 elements analyzed by PCR were absent from the orthologous positions in nonhuman primate genomes, except for a single element (L1HS72) that was also present in the common (Pan troglodytes) and pygmy (P. paniscus) chimpanzee genomes. Sequence analysis revealed that this single exception is the product of a gene conversion event involving an older preexisting L1 element. One hundred fifteen (45%) of the Ta L1 elements were polymorphic with respect to insertion presence or absence and will serve as identical-by-descent markers for the study of human evolution.

Figure 1

Human diversity associated with a truncated Ta L1Hs element, as shown by an agarose gel chromatograph of the PCR products from a survey of the human genomic variation associated with L1HS7. Amplification of the pre-integration site of this locus generates a 130-bp PCR product; amplification of a filled site generates a 326-bp product (by use of flanking unique sequence primers). In this survey of human genomic variation, 20 individuals from each of four diverse populations were assayed for the presence or absence of the L1 element, with only the African American samples shown here; the control samples (gray lines) were TLE buffer (i.e., 10 mM Tris-HCl:0.1 mM EDTA), common chimpanzee, gorilla, and owl monkey DNA templates. Most of the individuals surveyed were homozygous for the presence of the L1 element; in addition, this particular L1 element was absent from the genomes of nonhuman primates.

Figure 2

Human diversity associated with a long L1Hs Ta insertion polymorphism, as shown by an agarose gel chromatograph of the PCR products from a survey of the human genomic variation associated with L1HS364. Because of the size (∼6,000 bp) of this L1 element, two separate PCRs are performed to genotype individual samples. In the first reaction, flanking unique sequence primers were used to genotype the empty alleles (A); amplification of empty alleles from this locus generates a 97-bp PCR product. In the second reaction, a Ta subfamily–specific internal primer termed “ACA” and the 3′ flanking unique sequence primer were used to genotype filled sites (B); the amplification of filled sites generates a 170-bp product. In this survey of human genomic variation, 20 individuals from each of four diverse populations were assayed for the presence or absence of the L1 element, with only the Egyptian samples shown here; the control samples (black lines) were TLE buffer, common chimpanzee, gorilla, and owl monkey DNA templates. This particular L1 insertion polymorphism is a high-frequency insertion polymorphism, and most of the individuals surveyed have L1 filled chromosomes.

Figure 3

L1HS72 gene conversion. A, Agarose gel chromatograph of the PCR products derived from the amplification of L1HS72 in a series of human and nonhuman primate genomes, with a schematic of the primate evolutionary tree over the past 35 million years shown below. The yellow notched arrow represents the approximate time period when the L1HS72 element first integrated, and the red notched arrow represents the approximate time period of the gene conversion event of the preexisting L1 element. The fragment-length marker is a 123-bp ladder. B, Sequence alignment generated by sequencing the L1HS72 amplicons from nine diverse humans. Sequences are compared relative to L1Hs Ta consensus sequence and the L1HS72 sequence obtained from GenBank with only the diagnostic bases shown and positions reported relative to L1 retrotransposable element–1 (Dombroski et al. 1991). The G and C at positions 5536 and 5539 are indicative of the Ta-0 subset, whereas the Ta-1 subset has T and G at these nucleotides (Boissinot et al. 2000). The G at position 6015 (in addition to the ACA at positions 5930–5932) is diagnostic for the L1Hs Ta subfamily (Ovchinnikov et al. 2001). The target-site duplication sequence (TSD) is shown in brackets. The mosaic elements seen in the human samples are believed to be the result of at least one gene conversion, some time after the divergence of humans from the great apes (approximately five million years ago), of a preexisting L1 element with a younger L1Hs element. In the representation of nucleotides, different colors are used to denote conserved sequences and sequence variations between samples: green denotes bases unique to the common and pygmy chimpanzee genomes; blue denotes nucleotides unique to the human samples; orange denotes shared bases conserved between the common chimpanzee, pygmy chimpanzee, and human samples; and red denotes SNPs, within L1HS72, in the human population.

Figure 4

Ta L1 element size classes (in bp), showing the size distribution of Ta L1Hs elements. Elements are grouped in 500-bp intervals ranging from <500 bp to 7,000 bp long. The two most common size intervals are shown in black.

Figure 5

L1HS169-mediated transduction, showing an L1Hs transduction event. L1HS169 marked by clear target-site duplications is the putative source gene for L1HS28. The L1HS28 insertion contains 3′ flanking sequences identical to that of L1HS169 and unique target-site duplications flanking this entire sequence—suggesting that L1HS28 was created from a read-through transcript of L1HS169 that, to give rise to L1HS28, integrated into a new location on chromosome X. In addition, a second transduction event—L1HS547, from chromosome 18—is also flanked by unique target-site duplications and was also derived from L1HS169.