Humans and chimpanzees diverged approximately 6 million years ago, making the chimpanzee the closest extant relative to modern humans. The characterization of sequence changes both at the nucleotide and the structural level is therefore important for the understanding of primate evolution, including human-specific traits. At the nucleotide level, the identity of the genomes has been estimated to be 98% to 99% [1
], excluding insertions and deletions and other small rearrangements. The chimpanzee Chromosome 22 (PTR22), which is orthologous to human Chromosome 21(HSA21), was the first to be sequenced and the majority of the rest of the genome is represented as a draft assembly [2
]. The exact nucleotide substitution rate for the alignment of these sequences is 1.23% (excluding insertions and deletions)[2
]. Taking insertion and deletion events into account, the sequence identity has been estimated to be about 95% [7
In addition to nucleotide level changes, large structural rearrangements have also occurred between the species and they are discernable through comparison of the G-banded karyotypes. The most obvious structural difference between the human and chimpanzee genomes is the fusion of two acrocentric chromosomes creating human Chromosome 2. This results in a lower total chromosome number in humans (22, XY versus 23, XY). In addition, there are nine visible pericentric inversions affecting Chromosomes 1, 4, 5, 9, 12, 15, 16, 17, and 18 [8
]. Of these rearrangements, only the fusion creating Chromosome 2 and the inversions on Chromosome 1 and 18 are specific to the human lineage, while the remaining changes have occurred in the chimpanzee lineage.
Early comparative studies between the human and chimpanzee genomes focused mainly on localized sequencing efforts and characterization of karyotypically visible chromosomal rearrangements. More recently, a number of studies have been performed with the goal of characterizing loss and gain of submicroscopic regions of DNA using comparative genomic hybridization [9
]. The results reveal that copy number differences are abundant between the human and chimpanzee genomes, which agree with studies of segmental duplications in the genomes of several species [11
]. These latter studies show that there is a higher incidence of segmental duplications in the human genome than in the mouse or rat genomes, indicating that increases in sequence copy number are more common in recent primate evolution [14
]. The high frequency of copy number differences between humans and chimpanzees are also consistent with the findings that these types of structural variants are present as a common type of polymorphism in the human genome [15
Although recent technological advances allow for detection of most types of genomic variation, limitations in available methodology have prevented the genome-wide discovery of balanced rearrangements such as inversions. Nonetheless, the fact that nine known cytogenetically visible inversion events distinguish the human and chimpanzee genomes indicates that these may have been a common form of structural rearrangement during primate evolution. The comparative study of human Chromosome 21 and chimpanzee Chromosome 22 did not assess the extent of inversion events between the two species [6
]. The recent publication of the chimpanzee genome and accompanying comparative analysis of structural rearrangements did not address inversion events beyond those that are visible in the karyotype [2
Characterization of inversion events between humans and chimpanzees are important because inversions can affect the expression of genes adjacent to the breakpoints, or directly interrupt genes spanning the breakpoints. Large inversions have also been proposed to be a direct driving force in speciation [20
] and have been shown to suppress recombination [21
]. It is also important to investigate inversion events between humans and chimpanzees as the frequency of such events can provide an indication as to what extent inversion variants exist as polymorphisms in the human population. Since inversion polymorphisms are difficult to detect, there has not been, until very recently, an estimate of their occurrence in the human genome. By mapping fosmid ends to the reference genome sequence, Tuzun et al. identified 56 putative inversion breakpoints in a single individual (inversion breakpoint pairs cannot be identified unambiguously using this approach). This implies that inversion polymorphisms are much more common than previously assumed.
Using the draft sequence of Pan troglodytes (chimpanzee), we have used alignments between the human and chimpanzee genomes to identify regions of inverted orientation. Through this computational approach, there is no theoretical limit to the size resolution of inversion regions that can be identified and resolved. It is, therefore, possible to equally identify cytogenetically detectable, as well as nucleotide level inversions, with a resolution down to the breakpoint sequence itself. Of the 1,576 computationally predicted inversions, 23 have now been confirmed experimentally. Screening in human control individuals also revealed three regions to be polymorphic in the human population. Our data indicate that inversions have occurred frequently in recent primate evolution, and both computational analysis and experimental data support the observation that inversion polymorphisms may be common in the human genome.