A Comparative Analysis of numt Evolution in Human and Chimpanzee (original) (raw)

Journal Article

,

*Department of Zoology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Ramat Aviv 69978, Israel

Search for other works by this author on:

*Department of Zoology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Ramat Aviv 69978, Israel

†Department of Biology and Biochemistry, University of Houston

Search for other works by this author on:

1

Present address: National Evolutionary Synthesis Center, Durham, North Carolina, USA

William Martin, Associate Editor

Author Notes

Accepted:

11 October 2006

Published:

20 October 2006

• PDF
• Split View
• • Article contents
  • Figures & tables
  • Video
  • Audio
  • Supplementary Data
• Cite

Cite

Einat Hazkani-Covo, Dan Graur, A Comparative Analysis of numt Evolution in Human and Chimpanzee, Molecular Biology and Evolution, Volume 24, Issue 1, January 2007, Pages 13–18, https://doi.org/10.1093/molbev/msl149
Close

Navbar Search Filter Mobile Enter search term Search

Abstract

Mitochondrial DNA sequences are frequently transferred into the nuclear genome, giving rise to _numt_s (nuclear DNA sequences of mitochondrial origin). So far, the evolutionary history of _numt_s has largely been studied by using single genomes. Here, we present the first attempt to study numt evolution in a comparative manner by using a pairwise genomic alignment. The total number of _numt_s was estimated to be 452 in human and 469 in chimpanzee. numts that were found in both genomes at identical loci were deemed to be orthologous; 391 numts (>80%) were classified as such. The preponderance of orthologous _numt_s is due to the very short divergence time between the 2 hominoids. The rest of numts were deemed to be nonorthologous. Nonorthologous _numt_s were subdivided into 1) ancestral _numt_s that have lost an ortholog in one species through deletion (12 in human and 11 in chimpanzee), 2) new _numt_s acquired by the insertion of a mitochondrial sequence after the divergence of the 2 species (34 in human and 46 in chimpanzee), and 3) paralogous _numt_s created by the tandem duplication of a preexisting numt (2 in human). This approach also enabled us to reconstruct the numt repertoire in the common ancestor of humans and chimpanzees (409 numts). Our comparative approach is also useful in identifying the exact boundaries of numts.

Mitochondrial DNA sequences are frequently transferred into the nuclear genome, giving rise to _numt_s (nuclear DNA sequences of mitochondrial origin, Lopez et al. 1994). _numt_s have been described in more than 80 species (Bensasson et al. 2001). For most species, the estimate of numt content and abundance is still incomplete. However, with fully sequenced genomes, it is possible to obtain an accurate estimate of numt abundance (Richly and Leister 2004). There is no correlation between the fraction of noncoding DNA and numt abundance (Richly and Leister 2004). The reason for the variation in numt abundance among genomes is not known. Conceptually, the differences might be due to 1) different rates of numt insertion, 2) different rates of numt deletion, and 3) different rates of numt postinsertional duplication.

All mammalian numt studied to date were found to be functionless, and it is thought that they became pseudogenized on arrival into the nucleus because of the differences between the nuclear and mitochondrial genetic codes (Gellissen and Michaelis 1987; Perna and Kocher 1996). In yeast, _numt_s are transferred under natural conditions during the repair of double-strand breaks (Ricchetti et al. 1999), and it was suggested that this is the cause for the ongoing colonization of different genomes by _numt_s. The continuing process of numt integration into the nuclear genome is evidenced by the finding of _numt_s that have been inserted into the human genome after the human–chimpanzee divergence (Ricchetti et al. 2004). Some of these numts are variable with respect to genomic presence or absence, indicating that they have only arisen recently in the human population. Transposition of _numt_s into genes has also been associated with human diseases (Willett-Brozick et al. 2001; Turner et al. 2003; Goldin et al. 2004).

From human genome data, different estimates of the number of _numt_s have been put forward in the literature (Mourier et al. 2001; Tourmen et al. 2002; Woischnik and Moraes 2002; Bensasson et al. 2003; Richly and Leister 2004). Additionally, phylogenetic methods have been suggested for dating the insertion of _numt_s into the nuclear genome (Mourier et al. 2001; Woischnik and Moraes 2002). Initial results indicated a fairly rapid process of numt insertion, however, some studies ignored the possibility of postinsertional nuclear duplication (e.g., Bensasson et al. 2000) resulting in overestimation of numt insertion rates. Hazkani-Covo et al. (2003) suggested a methodology for dating the insertion of _numt_s into the nuclear genome by using a single nuclear genome sequence and a mitochondrial phylogenetic tree. This methodology had the advantage of being able to detect numt duplication events. We discovered that the rate of numt insertion on the branch leading to humans was much lower than previously reported (Mourier et al. 2001; Woischnik and Moraes 2002). Most _numt_s turned out to be paralogs of preexisting _numt_s, rather than new insertions.

Two _numt_s are defined as orthologous if they are derived from a speciation event, but as paralogous if they are derived from a duplication event. So far, the evolutionary history of _numt_s has largely been studied by means of paralogous comparisons within single genomes (Mourier et al. 2001; Woischnik and Moraes 2002; Hazkani-Covo et al. 2003). The availability of closely related completely sequenced genomes has enabled us to use comparative methods to study directly orthologous numt evolution. We note that by using the methodology of Hazkani-Covo et al. (2003), the existence of orthologous numt in species other than humans was inferred indirectly. That inference, however, yielded a testable prediction. Thus, for example, a numt that was inferred to have been inserted in the common ancestor of human and chimpanzee should possess orthologs in both species. However, this prediction could be wrong if the mitochondrial phylogenetic tree is not the true tree. In addition, this methodology is only applicable to long _numt_s that have sufficient phylogenetic signal. With 2 or more genomes, the presence of orthologous _numt_s can be inferred directly, even when the _numt_s are short.

In the following, we suggest a protocol based on genome alignment to estimate the number of _numt_s in closely related species. We apply this approach to the genomes of human (Lander et al. 2001) and chimpanzee (Pan troglodytes; Mikkelsen et al. 2005), and use the alignment to identify evolutionary events that may have affected numt composition in each genome, as well as to reconstruct the numt makeup in the common ancestor of human and chimpanzee.

Because there are no hot spots for numt insertion (Zischler 2000), the presence of a numt at a particular locus in both genomes was taken to imply orthology (fig. 1). Nonorthologous _numt_s that are present in only one genome are further classified into insertions, partial or total deletions, or tandem duplications (fig. 1). Each such event can take place in either lineage. Nonorthologous _numt_s are identified by a gap in the alignment. The distinction between insertions and deletions is based on the fact that there exists no known mechanism for the precise excision of _numt_s. Thus, if the gap coincides precisely with the boundaries of the numt, an insertion is inferred. If the gap is smaller or larger than the numt in the other genome, we infer the occurrence of a partial or total deletion, respectively. Tandem numt duplications are characterized by adjacent homologous numts and a gap coinciding perfectly with the boundaries of the homolog from the other species. The assumptions used for numt classifications here were also used in PCR-based numt recognition (e.g., Lopez et al. 1994; Zischler et al. 1998; Herrnstadt et al. 1999).

FIG. 1.—

(A) numt classification based on genome alignment of homologous loci between human and chimpanzee. (B) Each evolutionary event is positioned on the inferred branch on the phylogenetic tree.

Our analyses were based on genomic sequences and annotations from the University of California at Santa Cruz (Karolchik et al. 2004) Genome Center. First, Blast was used to search each of the human and chimpanzee genomes for regions of similarity with conspecific mitochondrial sequences (fig. 2, frame 1). Closely spaced mitochondrial hits were concatenated (fig. 2, frame 2). The distinction between orthologous and nonorthologous _numt_s (as described in fig. 1) was accomplished through a comparison of human and chimpanzee numt preliminary datasets. The comparison was based on the University of California-Santa Cruz genome alignment between human and chimpanzee. The analysis was performed in a reciprocal manner: comparing the human genome to the chimpanzee genome and comparing the chimpanzee genome to the human genome. For a detailed description of the methodology, see Supplementary Material online.

FIG. 2.—

Flowchart of data collection and numt classification in human. Two types of UCSC files were used in the analysis: the nucleotide pairwise alignment file and the alignment net file. The final numt classification is determined after comparison with the chimpanzee genome (for details see Supplementary Methods, Supplementary Material online).

We found a similar number of numts in both genomes: 452 _numt_s in human and 469 _numt_s in chimpanzee (table 1). The total number of _numt_s in the 2 genomes was found to be similar to previous estimates in the literature. Unsurprisingly, because of the short time that has passed since the divergence of the 2 hominoids, 391 numts (87% in human and 84% in chimpanzee) were classified as orthologous, that is, were inserted into the nuclear genome before the divergence between the 2 lineages (table S1 in Supplementary Material online).

Table 1

Numbers and Total Sizes (in Parentheses) of Different numt Types within the Genomes of Human and Chimpanzee

a

The total size of orthologous numts in chimpanzee was calculated according to human coordinates (see Supplementary Methods, Supplementary Material online). In order not to run into the risk of classifying the same numt twice, the size of partially deleted numts and very small tandem duplications (whose number appears in parentheses) was added to orthologous size. In addition, tandem duplications and evidence for partial deletion are not counted in the total number of numts. Underlined numts were used to estimate the repertoire of the common ancestor.

b

Human numts are listed as evidence for deletions in the chimpanzee genome; chimpanzee numts are listed as evidence for deletions in the human genome.

Table 1

Numbers and Total Sizes (in Parentheses) of Different numt Types within the Genomes of Human and Chimpanzee

a

The total size of orthologous numts in chimpanzee was calculated according to human coordinates (see Supplementary Methods, Supplementary Material online). In order not to run into the risk of classifying the same numt twice, the size of partially deleted numts and very small tandem duplications (whose number appears in parentheses) was added to orthologous size. In addition, tandem duplications and evidence for partial deletion are not counted in the total number of numts. Underlined numts were used to estimate the repertoire of the common ancestor.

b

Human numts are listed as evidence for deletions in the chimpanzee genome; chimpanzee numts are listed as evidence for deletions in the human genome.

We identified 46 previously undescribed postspeciation _numt_s in the chimpanzee. These ranged in size between 37 and 3,076 bp. In addition, we identified 34 _numt_s in human. Our study, thus, increases the number of known human-specific numts (Ricchetti et al. 2004) by 26%, and identifies the shortest (29 vs. 47 bp) and the longest (5,219 vs. 1,323 bp) new _numt_s. Human and chimpanzee postspeciation numts that were found in this study are listed in table 2. The common ancestor of human and chimpanzee is estimated to have lived about 6 Myr ago (Goodman et al. 1998). Thus, the average rates of numt insertion are 5.7 insertions per 1 Myr in human and 7.7 numt insertions per 1 Myr in chimpanzee. The difference is not statistically significant (P < 0.179).

Table 2

Postspeciation (New) _numt_s in the Human and Chimpanzee Genomes. Coordinates of the numts within the Chromosomes and the Mitochondria Are Shown as well as the numt size. Chromosome Names That Contain “Random” Include Unmapped Sequences from the Chromosome

Table 2

Postspeciation (New) _numt_s in the Human and Chimpanzee Genomes. Coordinates of the numts within the Chromosomes and the Mitochondria Are Shown as well as the numt size. Chromosome Names That Contain “Random” Include Unmapped Sequences from the Chromosome

From among the postspeciation _numt_s, only 2 cases of tandem duplication were found (both in the human genome). In the first case, an internal segment of 30 bp within a numt located on chromosome 10 was duplicated once. The second case, in chromosome 12, includes 18 tandem duplications of a 47-bp sequence (fig. 3).

FIG. 3.—

Multiple sequence alignment of 18 tandemly repeated _numt_s in human chromosome 12 (positions 125, 420, 954–125, 422, 037) and the homologous locus on chimpanzee chromosome 10. The alignment to human and chimpanzee mitochondria is also shown. Each repeat is 47 bp in length and aligns to mitochondrial coordinates 4418–4464 (box). The flanking regions of the human internally repeated numt align to human mitochondrial coordinates 4478–4382 and can also be aligned to a single chimpanzee numt. Duplications (Dup_) are numbered in order of their appearance from 5′ to 3′. Identical nucleotides in the alignment columns are indicated by a dot; dashes indicate gaps. Hs, Homo sapiens; Pt, Pan troglodytes.

The number of events in which _numt_s were deleted from the genome is fairly similar between the 2 species. There are 12 deletion events in human, of which 11 are total deletions and 1 is a partial one. In chimpanzee, there are 11 deletion events, of which 7 are total deletions and 4 are partial. As far as the total deletions are concerned, one can distinguish between 2 separate groups: most of the _numt_s seem to have been deleted from the genome as part of a much larger segment. However, in a few cases, the numt deletion included only a limited flanking region.

The number of nonorthologous _numt_s is not large enough to be able to detect differences in numt evolutionary dynamics (insertion, deletion, or tandem duplication) between the 2 lineages. Still, we are now able to reconstruct the _numt_s constitution in the common ancestor of the 2 hominoids. The number of _numt_s in the common ancestor of human and chimpanzee is estimated at 409. This number includes 391 _numt_s that are still found in the 2 genomes, and a total of 18 _numt_s that were lost from 1 of the 2 genomes. Given the very low rate of numt deletion, the possibility that a numt has been lost in both genomes seems negligible.

We suggest that in comparison to single genome analyses, our methodology resulted not only in a more accurate estimate of the number of numts but also in a more precise identification of their boundaries. First, this protocol distinguishes between orthologous and nonorthologous _numt_s. Second, by using genome alignment, we identified orthologous _numt_s that escaped detection by the usual Blasting of mitochondrial sequences against the nuclear genome. In 145 out of 391 cases, _numt_s were identified in only one of the genomes when the Blast analysis was used. However, in the majority of cases, alignment of those _numt_s to the corresponding fragment in the second genome revealed a cryptic or quasi-cryptic ortholog. In 15 cases, the existence of orthologous _numt_s in chimpanzee was inferred on the basis of a small stretch of Ns similar in size to the human numt in the homologous position. Finally, our protocol enables a more precise identification of the genomic coordinates of _numt_s. The comparative method allows concatenation of fragments that may otherwise be identified as independent numts.

We thank Shay Covo and Tal Dagan for their help. This work was supported in part by a grant (DBI-0543342) from the National Science Foundation.

References

Rates of DNA duplication and mitochondrial DNA insertion in the human genome

,

J Mol Evol

,

2003

, vol.

57

(pg.

343

-

354

)

Mitochondrial pseudogenes: evolution's misplaced witnesses

,

Trends Ecol Evol

,

2001

, vol.

16

(pg.

314

-

321

)

Frequent assimilation of mitochondrial DNA by grasshopper nuclear genomes

,

Mol Biol Evol

,

2000

, vol.

17

(pg.

406

-

415

)

Gene transfer. Mitochondria to nucleus

,

Ann N Y Acad Sci

,

1987

, vol.

503

(pg.

391

-

401

)

Transfer of a mitochondrial DNA fragment to MCOLN1 causes an inherited case of mucolipidosis IV

,

Hum Mutat

,

2004

, vol.

24

(pg.

460

-

465

)

Toward a phylogenetic classification of primates based on DNA evidence complemented by fossil evidence

,

Mol Phylogenet Evol

,

1998

, vol.

9

(pg.

585

-

598

)

Evolutionary dynamics of large numts in the human genome: rarity of independent insertions and abundance of post-insertion duplications

,

J Mol Evol

,

2003

, vol.

56

(pg.

169

-

174

)

A novel mitochondrial DNA-like sequence in the human nuclear genome

,

Genomics

,

1999

, vol.

60

(pg.

67

-

77

)

The UCSC Table Browser data retrieval tool

,

Nucleic Acids Res

,

2004

, vol.

32

(pg.

493

-

496

)

et al. ,

(256 co-authors)

.

Initial sequencing and analysis of the human genome

,

Nature

,

2001

, vol.

409

(pg.

860

-

921

)

Numt, a recent transfer and tandem amplification of mitochondrial DNA to the nuclear genome of the domestic cat

,

J Mol Evol

,

1994

, vol.

39

(pg.

174

-

190

)

et al. ,

(67 co-authors)

.

Initial sequence of the chimpanzee genome and comparison with the human genome

,

Nature

,

2005

, vol.

437

(pg.

69

-

87

)

The human genome project reveals a continuous transfer of large mitochondrial fragments to the nucleus

,

Mol Biol Evol

,

2001

, vol.

18

(pg.

1833

-

1837

)

Mitochondrial DNA: molecular fossils in the nucleus

,

Curr Biol

,

1996

, vol.

6

(pg.

128

-

129

)

Mitochondrial DNA repairs double-strand breaks in yeast chromosomes

,

Nature

,

1999

, vol.

402

(pg.

96

-

100

)

Continued colonization of the human genome by mitochondrial DNA

,

PLoS Biol

,

2004

, vol.

2

pg.

E273

NUMTs in sequenced eukaryotic genomes

,

Mol Biol Evol

,

2004

, vol.

21

(pg.

1081

-

1084

)

Structure and chromosomal distribution of human mitochondrial pseudogenes

,

Genomics

,

2002

, vol.

80

(pg.

71

-

77

)

Human genetic disease caused by de novo mitochondrial-nuclear DNA transfer

,

Hum Genet

,

2003

, vol.

112

(pg.

303

-

309

)

Germ line insertion of mtDNA at the breakpoint junction of a reciprocal constitutional translocation

,

Hum Genet

,

2001

, vol.

109

(pg.

216

-

223

)

Pattern of organization of human mitochondrial pseudogenes in the nuclear genome

,

Genome Res

,

2002

, vol.

12

(pg.

885

-

893

)

Nuclear integrations of mitochondrial DNA in primates: inference of associated mutational events

,

Electrophoresis

,

2000

, vol.

21

(pg.

531

-

536

)

A hominoid-specific nuclear insertion of the mitochondrial D-loop: implications for reconstructing ancestral mitochondrial sequences

,

Mol Biol Evol

,

1998

, vol.

15

(pg.

463

-

469

)

Author notes

1

Present address: National Evolutionary Synthesis Center, Durham, North Carolina, USA

William Martin, Associate Editor

© The Author 2006. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

Supplementary data

Citations

Views

Altmetric

Metrics

Total Views 3,995

3,466 Pageviews

529 PDF Downloads

Since 11/1/2016

Citations

81 Web of Science

×

Email alerts

Email alerts

Citing articles via

• Latest

• Most Read

• Most Cited

More from Oxford Academic