Birth and Evolutionary History of a Human Minisatellite
http://www.100md.com
分子生物学进展 2004年第2期
Departamento de Bioquímica y Biología Molecular, Facultad de Biología, Universidad de Santiago, Galicia, Spain
E-mail: bnjgm@usc.es.
Abstract
One of the most exciting challenges in human biology is the understanding of how our genome was constructed during evolution. Here we explore the evolutionary history of the low polymorphic human minisatellite MsH42 and its flanking sequences. We show that the evolutionary birth of MsH42 took place within an intron, early in primate lineage evolution, more than 40 MYA. Then, single base-pair changes and duplications/deletions of repeat blocks by mispairing were probably the main forces governing the generation of this minisatellite and its polymorphism throughout primate evolution. Moreover, we detected several phylogenetic footprints at both sides of MsH42. We believe that our findings will contribute to the understanding of low-variability minisatellite evolution.
Key Words: minisatellite evolution ? phylogenetic footprint ? primates ? human minisatellite MsH42
Introduction
Tandemly repeated DNA sequences are interspersed throughout the genomes of all eukaryotes, including humans (Charlesworth, Sniegowski, and Stephan 1994). Among those sequences, the minisatellites are regions of DNA characterized by the repetition of short sequence units in a tandem array, which can show multiallelic variation and high degrees of heterozygosity (Jeffreys 1987; Vergnaud and Denoeud 2000). Minisatellites are associated with important genomic processes such as gene regulation, imprinting, and recombination (Vergnaud and Denoeud 2000). Several mechanisms have been proposed to explain minisatellite variability, including replication slippage, unequal crossover, single base-pair mutations, gene conversion, and insertion/deletions (Olsen 1999; Vergnaud and Denoeud 2000), but the origin and evolutionary dynamics of such loci remain obscure. There is one study about the evolutionary transience of two hypervariable minisatellites loci showing that large fluctuations in the degree of polymorphism can occur over remarkably short periods of time (Gray and Jeffreys 1991). However, little is known about the long-term evolution of low-variability minisatellites.
The human minisatellite MsH42 (MsH42) is a low-polymorphic locus (three alleles have been identified in the human populations) that is localized in the chromosome 15q25.1 inside intron 5, between exons 5 and 6, of the gene Q9ULM1 (fig. 1A). This GC-rich minisatellite is able to interact specifically with nuclear proteins (Boán et al. 1997). Moreover, MsH42, with its proximal flanking sequences (the MsH42 region), is able to enhance in vitro intramolecular homologous recombination, promoting high rates of equal crossovers (Boán et al. 1998, 2002).
Table 1 MsH42 Alleles in Primates.
Materials and Methods
Sources and Preparation of Genomic DNA
For the present study, we used genomic DNA from 22 unrelated African humans as well as from 11 species of primates distributed as follows: seven chimpanzees (Pan troglodytes), six gorillas (Gorilla gorilla), four orangutans (Pongo pygmaeus), one gibbon (Hylobates lar), three macaques (one Macaca mulatta and two Macaca nigra), one mandrill (Mandrillus sphinx), one tamarin (Saguinus oedipus), and three lemurids (Lemur catta, Eulemur fulvus, and Varecia variegata). Genomic DNA from blood samples was prepared using the QIAamp DNA kit (QIAGEN) following the specifications of the manufacturer.
PCR Amplifications and DNA Sequencing
Primer pairs PS1/PS2 and P1/P2 were used to amplify the MsH42 region. Primer pairs E5.1/E5.2 and E6.1/E6.2 were used to amplify exons 5 and 6 of the Q9ULM1 gene (www.ensembl.org/Homo_sapiens), respectively. Primers PC1 and PC2 were designed based on two highly conserved motifs situated near MsH42 and were employed to amplify the minisatellite and its proximal flanking sequences. The sequences of the primers are PS1 (5'-CTGCAGCAATGGACTCAAAA-3'), PS2 (5'-CTGCAGACTCCAAATCCTAA-3'), P1 (5'-CTTGGGCACTCTAGGACACC-3'), P2 (5'-CACAGCTCTGGCTACAAGAG-3'), E5.1 (5'-TTTGCTCTGGGATTTAAGGC-3'), E5.2 (5'-CAACAAGCCATTGGAGCCAT-3'), E6.1 (5'-ATCAAGGACGTTGT GGGCTA-3'), E6.2 (5'-TTGCAG TCTTGCCTGGGCTT-3'), PC1 (5'-GGGCAGTGTTGAGAGTGAGC-3'), and PC2 (5'-TATCTTCATGAACTCACACT-3'). The general conditions for all amplifications and the cycling conditions for PS1/PS2 and P1/P2 were as described elsewhere (Boán et al. 1997, 2000). The denaturing, annealing, and extension steps for the new combinations of primers were PC1/PC2: 95°C 1 min, 56°C 40 s, and 72°C 30 s; E5.1/E5.2: 95°C 1 min, 58°C 15 s, and 72°C 20 s; E6.1/E6.2: 95°C 1 min, 60°C 15 s, and 72°C 20 s; E5.1/PS2: 95°C 1 min, 54°C 40 s, and 72°C 2 min. PCR products were cycle sequenced using the 377 DNA Automated Sequencer (Applied Biosystems).
Footprinting Analysis
The probes for footprinting experiments were the Fp1 and Fp2 DNA fragments (fig. 1A). The Fp1 fragment was synthesized by PCR with primers PS1 and Pf1 (5'-GCCTCTCCCAGCTCTCCCAGCCCT-3') as follows: denaturing 95°C 1 min, annealing 70°C 30 s, and extension 72°C 50 s. The Fp2 fragment was obtained by DdeI digestion of the P1/P2 amplification fragment of the MsH42 region. These fragments were cloned in the pGemT-easy vector (Promega Inc.) in both orientations, and the probes were obtained by digestion with SacII and NotI. The NotI end of both strands (upper and lower) was end-labeled with radioactive nucleotides yielding specific activities of 5 x 108 cpm/μg. The nuclear extracts were prepared from testis, brain, and liver of 3-month-old Sprague-Dawley rats (Boán et al. 1998). The DNase I footprinting reactions were performed with the SureTrack Footprinting kit (Amersham Pharmacia Biotech) by mixing 50 μg of nuclear extracts, 3 μg of poly(dI-dC)·(dI-dC), and 0.5 μg of calf thymus DNA in the binding buffer with 0.1 ng of radioactive probe. The DNase I digestions were set up as follows: 0.1 U for 1.5 min in the reaction with BSA instead of protein extract (control reactions), 1.5 U for 1.5 min for reactions with testes extract, and 0.1 U for 1 min in experiments with brain or liver extracts. The purified DNAs were analyzed on 6% sequencing gels.
FIG. 1. The human MsH42 region and summary of PCR amplifications. (A) Scheme of the MsH42 region situated within intron 5 of the Q9ULM1 gene in chromosome 15q25.1. Arrows mark the position of the primers (E5.1, E5.2, PS1, P1, PC1, Pf1, PC2, P2, PS2, E6.1, and E6.2) employed in the amplifications. E5 = exon 5; E6 = exon 6. Fp1 and Fp2 represent the fragments used in the footprinting experiments. DdeI signals its restriction site. (B), PCR amplifications of primate DNAs with different primer pairs. The plus sign (+) denotes amplification and the minus sign (–) indicates no amplification. HUM = human; CHI = chimpanzee; GOR = gorilla; ORA = orangutan; GIB = gibbon; MAC = macaque; MDR = mandrill; TAM = tamarin; LCA = Lemur catta
Phylogenetic Analysis
The DNA sequences were aligned using the ClustalW program, employing the HUM(S), CHI1, and GOR1 alleles as the human, chimpanzee, and gorilla versions of the MsH42 region, respectively. The homogeneity of the sequence data sets was evaluated with the incongruence length difference test. The model that best fits the standard tree primate topology to the sequences was selected by the likelihood ratio test available in Modeltest version 3.0. Phylogenetic analysis are based on the neighbor-joining (distances were estimated under maximum-likelihood criteria), maximum-parsimony, and maximum-likelihood (GTR+G model) methods available in the PAUP* version 4 program (Swofford 1998). The node support was evaluated by bootstrapping using 1,000 resampled versions of the original data sets. All these analytical methods are reviewed by Thornton and DeSalle (2000). The likelihood values for the standard primate topology and the observed topologies were compared using the parametric Kishino-Hasegawa test (Kishino and Hasegawa 1989).
The GenBank accession numbers for the DNA sequences in this study are AY270192 to AY270201 for the minisatellite region, AY268962 to AY268968 for exon 5, and AY268969 to AY268975 for exon 6.
Results and Discussion
To establish whether the MsH42 region is present in primates, we carried out several PCR experiments using different combinations of primers (fig. 1). Primers PS1/PS2 amplified the genomic DNA from chimpanzee, gorilla, orangutan, and gibbon. However, in the case of gorilla DNA, the first round of PCR produced weak amplifications, and reamplification was therefore carried out on agarose-purified products. Attempts to amplify the genomic DNA of Old World monkeys (macaques and mandrill), New World monkeys (tamarin), and prosimian lemurids (Lemur, Eulemur, and Varecia) with those primers were unsuccessful. For this reason, we tried the amplification with primers P1/P2, which also span the minisatellite region. These primers gave successful amplification with the macaque and tamarin genomic DNA but failed to amplify the mandrill and lemurid DNAs, even when different combinations of the four primers were used. Mandrill genomic DNA was finally amplified with primers PC1/PC2, but once again the results with the prosimian DNA were negative. In table 1, the number of repeats, the presence of repeat variants, and the size of all primate MsH42-related alleles are summarized.
We performed amplifications with E5.1/PC2 to determine if the MsH42 region was also localized within an intron in primates. The results of these PCRs demonstrated the intronic localization of the MsH42-related minisatellite in primates (data not shown). On the other hand, we sought for the presence of the MsH42 region in the mouse, rat, zebrafish, and Drosophila genomic databases (www.ensembl.org). The result of these searches did not reveal any homologous sequences to the MsH42 region in those genomes. Because the genomes of both murine species contain the exons that flank MsH42, and there is no trace of the minisatellite, it is reasonable to conclude that the MsH42 minisatellite was originated during primate evolution.
To analyze the organization of the MsH42 region in the primates, the PCR products were sequenced. Figure 2 shows the alignment of a representative allele from each species in relation to the human MsH42 short allele. The tamarin version of the MsH42 locus has two repeat variants (A and C) common to humans, intercalated with other similar repeats, reminiscent of a pre-MsH42 region. In the Cercopithecoidea monkeys, macaque and mandrill, the minisatellite contains seven repeat units with the same arrangement in both species except for a G/A transition; this result agrees with the close phylogenetic relationship between these monkeys. Furthermore, in these primates, a constant characteristic of MsH42 emerges for the first time in all species: A is the first variant repeat and C1 is the last one (from 5' to 3'). In the gibbon DNA, the organization of the minisatellite is rather unusual and consists of 16 repeats, four of them restricted to this species (A5, C3, C4, D). The orangutan minisatellite has a repeat composition that is very similar to humans, and among its 13 repeats, only B4 is unique to this species. The panorama in the African great apes changes dramatically and the organization of MsH42 becomes highly homologous to the human locus. Thus, the analysis of gorilla samples revealed the existence of three alleles (table 1) with only one variant repeat (C2) absent in the human alleles. All gorilla alleles share a group of nine repeats at the 5' end of the minisatellite and another group of five repeats at the 3' end. Consequently, the differences among these alleles lie in the central portion of the minisatellite. The chimpanzees alleles (table 1) show an organization very similar to the human short allele, with the exception of two rare repeats (C2 and B3). It is worth noting that the GOR1 and CHI1 alleles are highly homologous to the human short allele, pointing towards their common origin from an ancestral array.
FIG. 2. Alignment among primate sequences of MsH42 region (5' flanking, 3' flanking, and minisatellite MsH42), exon 5, and exon 6. The human sequence is represented by the short allele, and chimpanzee and gorilla are represented by the CHI1 and GOR1 alleles, respectively. In the other primates, the only known allele is shown. To facilitate allele comparisons, the alignment of the minisatellite is shown using the one letter code (see table 1 for the key code). 5'HCS and 3'HCS boxes indicate the conserved sequences that include the PC1 and PC2 primers (underlined) and the 5f1 and 3f1 DNaseI footprints. chi boxes show the conservation of chi (GCTGGTGG) prokaryotic recombination hotspot. In the aligned sequences of exon 5, lined boxes mark two codons coding for methionine in almost all nonhuman primates and for threonine and valine in humans
Several conclusions can be drawn from the comparisons among primate alleles: (1) our findings agree with the belief that the African apes are more closely related to humans than the orangutans in Asia (Miyamoto, Slightom, and Goodman 1987; P??bo 2003); (2) the repeats A4 and B1 are exclusive of humans; (3) the repeats A and C probably are the most ancient; (4) the repeats A and C1 mark the 5' and 3' ends of the minisatellite, respectively; and (5) the existence of repeat blocks conserved in almost all species suggests that their presence could be necessary to construct this minisatellite.
Genomic sequence comparisons among distant species are a useful tool to identify regulatory elements present in the noncoding fraction of the genome. For example, sequence-specific conservation of noncoding DNA may imply functional constraint on these sequences and slower rates of molecular evolution (Ludwig 2002). Close examination of the sequences flanking MsH42 revealed an important conservation among primates (fig. 2). In particular, there are two highly conserved stretches (HCS) at both sides of the minisatellite, denoted as 5'HCS and 3'HCS. The 5'HCS comprises 29 bp with only two nucleotide changes in the tamarin, whereas the 3'HCS comprises 26 bp and shows only two transitions (A/G) in the macaque. It is likely that mandrill has both HCS because its genomic DNA was amplified with primers PC1/PC2, whose sequence is inside such regions. Phylogenetic footprinting (Tagle et al. 1988; Gumucio et al. 1992) and phylogenetic shadowing (Boffelli et al. 2003) have been used to identify putative regulatory elements, exploiting alignments across numerous distantly related or closely related species. We have previously detected protein-DNA interactions in this area by band-shifting experiments (Boán et al. 1997). Therefore, we carried out footprinting analysis to determine if these HCS were specifically recognized by proteins, thereby reflecting a function within the genome. For this purpose, we used the probes Fp1 and Fp2 (fig. 1). Fp1 contains the first 127 bp of the MsH42 region, including the 5'HCS and the first nine repeats. Fp2 has the last nine repeats of MsH42 and the next 175 bp downstream MsH42, including the 3'HCS. The results obtained with both orientations (upper and lower strands) of Fp1, using nuclear extracts from liver and brain, demonstrated the presence of two protected zones, 5f1 and 5f2, localized in the 5' flanking region of MsH42 (fig. 3). In the experiments performed with the Fp2 probe, we found two protections, 3f1 and 3f2, with the liver extract (fig. 3). Noteworthy, the 5f1 and 3f1 footprints include the 5'HCS and the 3'HCS, respectively, demonstrating the existence of a protein(s) that specifically interact with these conserved sequences, resembling phylogenetic footprints. Testes nuclear extract did not show any protected zone (fig. 3), indicating that testes may not express the protein(s) that recognizes such conserved zones. Taking into account that the MsH42 locus is situated within an intron, it is tempting to speculate that these HCS could play a role as regulatory elements of the Q9ULM1 gene. It should be mentioned that the footprinting experiments were done using rat nuclear extracts. Given that the rat genome does not contain the MsH42 region, it is reasonable to conclude that the sequences involved in the generation of footprints are present elsewhere in the rat genome and, hence, highly conserved.
FIG. 3. DNaseI footprinting analysis with the Fp1 and Fp2 probes. Upper strand corresponds to the sequence shown in figure 2, whereas lower strand is the complementary one. 5f1, 5f2, 3f1, and 3f2 indicate the DNaseI protected zones whose sequences are shown at the bottom. Arrows mark the start of MsH42 minisatellite. Lanes: G+A indicates Maxam and Gilbert G+A sequencing reactions; Control indicates reactions with BSA; Liver, Brain, Testes, indicates reactions with the corresponding nuclear extract
To compare the evolutionary dynamics of the intronic MsH42 region with respect to the proximal coding sequences, we amplified and sequenced the MsH42 neighbor exons 5 and 6 (fig. 1). The exon 5 amplified in all the primates studied, except in the lemurids, and the majority of changes were synonymous at the third codon position (fig. 2). We noticed that there is a methionine codon in all nonhuman primates that codes for threonine in humans, suppressing in this way a putative initiation codon. Perhaps this potential initiation codon might serve to produce more than one protein from the same mRNA by a mechanism involving translational initiation using internal ribosome binding sites (Oh and Sarnow 1993). Exon 6 is highly conserved in all primates except tamarin, all the changes being synonymous with respect to the human exon (fig. 2).
A phylogenetic analysis using the DNA sequences from the minisatellite region and exons 5 and 6 was performed. The results from the incongruence-length difference test indicate a significant character congruence (P = 1) that allows us to reconstruct a neighbor-joining topology from the total data set (data not shown). The phylogenetic tree so obtained is identical to the standard hypothesis about the phylogenetic relationships of primates (Miyamoto, Slightom, and Goodman 1987). Under the parsimony criterion, the total and the minisatellite flanking data sets also reflect the widely accepted primate tree, whereas alternative topologies were recovered from the exons and minisatellite sequences (data not shown). However, the consistency-index values as well as the Kishino-Hasegawa test indicate that these sequences make a consistent statement with respect to the standard primate topology. These results indicate that the mutational events observed in the minisatellite MsH42 are in agreement with the primate evolutionary history described for other noncoding sequences (Saitou and Ueda 1994; Apoil and Blancher 2000). Moreover, the estimation of parameter ( = 0.256) from population data suggests a mutation rate in the range of those estimated for other intron sequences in human genome (Huang et al. 1998).
The results presented here provide strong evidence that MsH42 was originated within an intron from a progenitor sequence, without a well-defined minisatellite structure, that experienced mutations leading to the formation of the first MsH42 repeat variants (fig. 4). Such progenitor sequence could have been originated by slipped-strand mispairing and unequal crossing-over between noncontiguous repeats formed by chance (Levinson and Gutman 1987; Haber and Louis 1998; Taylor and Breden 2000). The existence of the MsH42-like region in tamarin indicates its ancient generation during the earliest primate lineage evolution, before the divergence between Old World and New World monkeys about 40 MYA (Goodman 1999). This belief is strongly supported by the lack of amplifications in three prosimian species and because MsH42 homologous sequences could not be retrieved from the genomic databases of other organisms.
FIG. 4. Hypothetical evolutionary history of the minisatellite MsH42. The minisatellite organization of the different alleles for each species is shown. The repeat units are represented by a letter code (see table 1). It should be noted that the code was modified with respect to the previous one (Boán et al. 1998, 2002). Repeats present in human MsH42 are colored, and its probable appearance in the primate lineage is shown at the internodes along the tree. R in the common ancestor array indicates repeats of unknown sequence. Divergence times were taken from P??bo (2003) and Goodman (1999). Mys indicates million years ago
The homology between the human MsH42 and the corresponding locus in the African great apes, chimpanzee and gorilla, is remarkable. Comparison of the human short allele with the CHI1 and GOR1 alleles revealed an almost complete identity among the three sequences. There are two blocks of repeats, A-B-A-C-A1-B-A2-C-A and A-A-C-A-B2-A2-C-A-C-A-A3, identical in the three species, suggesting the existence of a common ancestral allelic array that has not become extinct in the actual human, chimpanzee, and gorilla populations. It is possible that the present polymorphism in humans and African great apes was originated from these highly homologous alleles. Surprisingly, the allele GOR1 is more similar to the human and chimpanzee alleles than to the other gorilla alleles (GOR2 and GOR3). This finding raises the possibility that several allelic forms of the MsH42 region coexisted in the ancestral genetic pool and that GOR2 and GOR3 alleles evolved from any of those forms, whereas GOR1, CHI1, and the HUM(S) alleles evolved from another allelic array. Nonetheless, the existence of these highly homologous sequences in the three species indicates that this ancestral common array should have been the most abundant allele. In fact, the identification of shared alleles between species rather than the relative frequencies is an accurate method to establish ancestral/derived status of polymorphic alleles (Iyengar et al. 1998).
The theory that modern humans originated in Africa is strongly supported by analysis of genetic variation in people today and from fossil discoveries (Stinger 2003). The existence of a highly homologous MsH42 allele in humans and African great apes, together with the fact that the frequency of the short allele in African humans (0.48 ± 0.07) is three times higher than the observed frequency in Europeans (0.16 ± 0.02) (Boán et al. 2002), supports the "Out of Africa" hypothesis. What is the origin of human MsH42 polymorphism? According to our results, the sequence of events that generated the present MsH42 polymorphism could have been as follows: First, there was the short allele coming from an evolutionary ancestor common to the great apes, second, a single duplication of a repeat block produced the long allele, third, the long allele generated the middle one by a deletion (fig. 4). The maintenance of the repeat arrangement in the three alleles points towards the mispairing of repeat blocks as the most probable mutational events in the human MsH42, as it has been proposed for many minisatellites (Charlesworth, Sniegowski, and Stephan 1994).
A useful approach to study human evolution at the molecular level is to consider our genome as a mosaic in which each DNA segment has its own evolutionary history (P??bo 2003). Our results allowed us to figure out when the human MsH42 began to exist early in primates, the birth of the minisatellite, and how this locus evolved to become MsH42 in humans. We believe that the present work is a worthwhile contribution to the knowledge of the low-variability minisatellites origin and that the evolutionary analysis of the MsH42 region is a small step toward the understanding of how the human genome has been made.
Acknowledgements
We thank M. Delclaux and coworkers from the Zoo of Madrid for providing us with the primate blood samples. We also thank J. Bertranpetit, A. Blancher, J. Bullerdiek, S. P??bo, P. Rogalla, W. Schempp, and R. Wimmer for giving us primate genomic DNA samples. Finally, we thank J. L. Caeiro for the African DNA samples. We would like to acknowledge J. M. Rodríguez for his help in some experiments and P. Barros for her help with figures. This project was funded by a grant of the Spanish Ministerio de Ciencia y Tecnología (BMC2001-3242). M.G.B. is recipient of a FPU grant from the Spanish government.
Literature Cited
Apoil, P.-A., and A. Blancher. 2000. Rh gene evolution in primates: study of intron sequences. Mol. Biol. Evol. 17:127-136.
Boán, F., A. González, J. M. Rodríguez, and J. Gómez-Márquez. 1997. Molecular characterization of a new human minisatellite that is able to form single-stranded loops in vitro and is recognized by nuclear proteins. FEBS Lett. 418:251-257.
Boán, F., J. M. Rodríguez, and J. Gómez-Márquez. 1998. A non-hypervariable human minisatellite strongly stimulates in vitro intramolecular homologous recombination. J. Mol. Biol. 278:499-505.
Boán, F., J. M. Rodríguez, S. Mouri?o, M. G. Blanco, A. Vi?as, L. Sánchez, and J. Gómez-Márquez. 2002. Recombination analysis of the human minisatellite MsH42 suggests the existence of two distinct pathways for initiation and resolution of recombination at MsH42 in rat testes nuclear extracts. Biochemistry 41:2166-2176.
Boffelli, D., J. McAuliffe, D. Ovcharenko, K. D. Lewis, I. Ovcharenko, L. Pachter, and E. M. Rubin. 2003. Phylogenetic shadowing of primate sequences to find functional regions of the human genome. Science 299:1391-1394.
Charlesworth, B., P. Sniegowski, and W. Stephan. 1994. The evolutionary dynamics of repetitive DNA in eukaryotes. Nature 371:215-220.
Goodman, M. 1999. The genomic record of humankind's evolutionary roots. Am. J. Hum. Genet. 64:31-39.
Gray, I. C., and J. A. Jeffreys. 1991. Evolutionary transience of hypervariable minisatellites in man and the primates. Proc. R. Soc. Lond. B Biol. Sci. 243:241-253.
Gumucio, D. L., H. Heilstedt-Williamson, T. A. Gray, S. A. Tarlé, D. A. Shelton, D. A. Tagle, J. L. Slightom, L. Goodman, and S. F. Collins. 1992. Phylogenetic footprinting reveals a nuclear protein which binds to silencer sequences in the human gamma and epsilon globin genes. Mol. Cell. Biol. 12:4919-4929.
Haber, J. E., and E. J. Louis. 1998. Minisatellite origins in yeast and humans. Genomics 48:132-135.
Huang, W., Y. X. Fu, B. H. Chang, X. Gu, L. B. Jorde, and W. H. Li. 1998. Sequence variation in ZFX introns in human populations. Mol. Biol. Evol. 15:138-142.
Iyengar, S., M. Marinero, C. Deinard, H. C. Rosenbaum, G. Sirugo, C. Castiglione, J. R. Kidd, and K. K. Kidd. 1998. Analyses of cross species polymerase chain reaction products to infer the ancestral state of human polymorphisms. DNA Seq. 8:317-327.
Jeffreys, A. J. 1987. Highly variable minisatellites and DNA fingerprints. Biochem. Soc. Trans. 15:309-317.
Kishino, H., and M. Hasegawa. 1989. Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data and the branching order in hominoidea. J. Mol. Evol. 29:170-179.
Levinson, G., and G. A. Gutman. 1987. Slipped-strand mispairing: a major mechanism for DNA sequence evolution. Mol. Biol. Evol. 4:203-221.
Ludwig, M. Z. 2002. Functional evolution of noncoding DNA. Curr. Opin. Genet. Dev. 12:634-639.
Miyamoto, M. M., J. L. Slightom., and M. Goodman. 1987. Phylogenetic relations of humans and African apes from DNA sequences in the -globin region. Science 238:369-373.
Oh, S., and P. Sarnow. 1993. Gene regulation: translational initiation by internal ribosomal binding. Curr. Opin. Genet. Dev. 3:295-300.
Olsen, K. M. 1999. Minisatellite variation in a single-copy nuclear gene: phylogenetic assessment of repeat length homoplasy and mutational mechanism. Mol. Biol. Evol. 16:1406-1409.
P??bo, S. 2003. The mosaic that is our genome. Nature 421:409-412.
Saitou, N., and S. Ueda. 1994. Evolutionary rates of insertion and deletion in noncoding nucleotide sequences of primates. Mol. Biol. Evol. 11:504-512.
Stinger, C. 2003. Human evolution: out of Ethiopia. Nature 423:692-694.
Swofford, D. L. 1998. PAUP*: phylogenetic analysis using parsimony (*and other methods). Version 4. Sinauer Associates, Sunderland, Mass.
Tagle, D. A., B. F. Koop, M. Goodman, J. L. Slightom, D. Hess, and R. T. Jones. 1988. Embryonic epsilon and gamma globin genes of a prosimian primate (Galago crassicaudatus): nucleotide and aminoacid sequences, developmental regulation and phylogenetic footprints. J. Mol. Biol. 203:439-455.
Taylor, J. S., and F. Breden. 2000. Slipped-strand mispairing at non-contiguous repeats in Poecilia reticulata: a model for minisatellite birth. Genetics 155:1313-1320.
Thornton, J. W., and R. DeSalle. 2000. Gene family evolution and homology: genomics meets phylogenetics. Annu. Rev. Genomics Hum. Genet. 1:41-73.
Vergnaud, G., and F. Denoeud. 2000. Minisatellites: mutability and genome architecture. Genome Res. 10:899-907.(Francisco Boán, Miguel G.)
E-mail: bnjgm@usc.es.
Abstract
One of the most exciting challenges in human biology is the understanding of how our genome was constructed during evolution. Here we explore the evolutionary history of the low polymorphic human minisatellite MsH42 and its flanking sequences. We show that the evolutionary birth of MsH42 took place within an intron, early in primate lineage evolution, more than 40 MYA. Then, single base-pair changes and duplications/deletions of repeat blocks by mispairing were probably the main forces governing the generation of this minisatellite and its polymorphism throughout primate evolution. Moreover, we detected several phylogenetic footprints at both sides of MsH42. We believe that our findings will contribute to the understanding of low-variability minisatellite evolution.
Key Words: minisatellite evolution ? phylogenetic footprint ? primates ? human minisatellite MsH42
Introduction
Tandemly repeated DNA sequences are interspersed throughout the genomes of all eukaryotes, including humans (Charlesworth, Sniegowski, and Stephan 1994). Among those sequences, the minisatellites are regions of DNA characterized by the repetition of short sequence units in a tandem array, which can show multiallelic variation and high degrees of heterozygosity (Jeffreys 1987; Vergnaud and Denoeud 2000). Minisatellites are associated with important genomic processes such as gene regulation, imprinting, and recombination (Vergnaud and Denoeud 2000). Several mechanisms have been proposed to explain minisatellite variability, including replication slippage, unequal crossover, single base-pair mutations, gene conversion, and insertion/deletions (Olsen 1999; Vergnaud and Denoeud 2000), but the origin and evolutionary dynamics of such loci remain obscure. There is one study about the evolutionary transience of two hypervariable minisatellites loci showing that large fluctuations in the degree of polymorphism can occur over remarkably short periods of time (Gray and Jeffreys 1991). However, little is known about the long-term evolution of low-variability minisatellites.
The human minisatellite MsH42 (MsH42) is a low-polymorphic locus (three alleles have been identified in the human populations) that is localized in the chromosome 15q25.1 inside intron 5, between exons 5 and 6, of the gene Q9ULM1 (fig. 1A). This GC-rich minisatellite is able to interact specifically with nuclear proteins (Boán et al. 1997). Moreover, MsH42, with its proximal flanking sequences (the MsH42 region), is able to enhance in vitro intramolecular homologous recombination, promoting high rates of equal crossovers (Boán et al. 1998, 2002).
Table 1 MsH42 Alleles in Primates.
Materials and Methods
Sources and Preparation of Genomic DNA
For the present study, we used genomic DNA from 22 unrelated African humans as well as from 11 species of primates distributed as follows: seven chimpanzees (Pan troglodytes), six gorillas (Gorilla gorilla), four orangutans (Pongo pygmaeus), one gibbon (Hylobates lar), three macaques (one Macaca mulatta and two Macaca nigra), one mandrill (Mandrillus sphinx), one tamarin (Saguinus oedipus), and three lemurids (Lemur catta, Eulemur fulvus, and Varecia variegata). Genomic DNA from blood samples was prepared using the QIAamp DNA kit (QIAGEN) following the specifications of the manufacturer.
PCR Amplifications and DNA Sequencing
Primer pairs PS1/PS2 and P1/P2 were used to amplify the MsH42 region. Primer pairs E5.1/E5.2 and E6.1/E6.2 were used to amplify exons 5 and 6 of the Q9ULM1 gene (www.ensembl.org/Homo_sapiens), respectively. Primers PC1 and PC2 were designed based on two highly conserved motifs situated near MsH42 and were employed to amplify the minisatellite and its proximal flanking sequences. The sequences of the primers are PS1 (5'-CTGCAGCAATGGACTCAAAA-3'), PS2 (5'-CTGCAGACTCCAAATCCTAA-3'), P1 (5'-CTTGGGCACTCTAGGACACC-3'), P2 (5'-CACAGCTCTGGCTACAAGAG-3'), E5.1 (5'-TTTGCTCTGGGATTTAAGGC-3'), E5.2 (5'-CAACAAGCCATTGGAGCCAT-3'), E6.1 (5'-ATCAAGGACGTTGT GGGCTA-3'), E6.2 (5'-TTGCAG TCTTGCCTGGGCTT-3'), PC1 (5'-GGGCAGTGTTGAGAGTGAGC-3'), and PC2 (5'-TATCTTCATGAACTCACACT-3'). The general conditions for all amplifications and the cycling conditions for PS1/PS2 and P1/P2 were as described elsewhere (Boán et al. 1997, 2000). The denaturing, annealing, and extension steps for the new combinations of primers were PC1/PC2: 95°C 1 min, 56°C 40 s, and 72°C 30 s; E5.1/E5.2: 95°C 1 min, 58°C 15 s, and 72°C 20 s; E6.1/E6.2: 95°C 1 min, 60°C 15 s, and 72°C 20 s; E5.1/PS2: 95°C 1 min, 54°C 40 s, and 72°C 2 min. PCR products were cycle sequenced using the 377 DNA Automated Sequencer (Applied Biosystems).
Footprinting Analysis
The probes for footprinting experiments were the Fp1 and Fp2 DNA fragments (fig. 1A). The Fp1 fragment was synthesized by PCR with primers PS1 and Pf1 (5'-GCCTCTCCCAGCTCTCCCAGCCCT-3') as follows: denaturing 95°C 1 min, annealing 70°C 30 s, and extension 72°C 50 s. The Fp2 fragment was obtained by DdeI digestion of the P1/P2 amplification fragment of the MsH42 region. These fragments were cloned in the pGemT-easy vector (Promega Inc.) in both orientations, and the probes were obtained by digestion with SacII and NotI. The NotI end of both strands (upper and lower) was end-labeled with radioactive nucleotides yielding specific activities of 5 x 108 cpm/μg. The nuclear extracts were prepared from testis, brain, and liver of 3-month-old Sprague-Dawley rats (Boán et al. 1998). The DNase I footprinting reactions were performed with the SureTrack Footprinting kit (Amersham Pharmacia Biotech) by mixing 50 μg of nuclear extracts, 3 μg of poly(dI-dC)·(dI-dC), and 0.5 μg of calf thymus DNA in the binding buffer with 0.1 ng of radioactive probe. The DNase I digestions were set up as follows: 0.1 U for 1.5 min in the reaction with BSA instead of protein extract (control reactions), 1.5 U for 1.5 min for reactions with testes extract, and 0.1 U for 1 min in experiments with brain or liver extracts. The purified DNAs were analyzed on 6% sequencing gels.
FIG. 1. The human MsH42 region and summary of PCR amplifications. (A) Scheme of the MsH42 region situated within intron 5 of the Q9ULM1 gene in chromosome 15q25.1. Arrows mark the position of the primers (E5.1, E5.2, PS1, P1, PC1, Pf1, PC2, P2, PS2, E6.1, and E6.2) employed in the amplifications. E5 = exon 5; E6 = exon 6. Fp1 and Fp2 represent the fragments used in the footprinting experiments. DdeI signals its restriction site. (B), PCR amplifications of primate DNAs with different primer pairs. The plus sign (+) denotes amplification and the minus sign (–) indicates no amplification. HUM = human; CHI = chimpanzee; GOR = gorilla; ORA = orangutan; GIB = gibbon; MAC = macaque; MDR = mandrill; TAM = tamarin; LCA = Lemur catta
Phylogenetic Analysis
The DNA sequences were aligned using the ClustalW program, employing the HUM(S), CHI1, and GOR1 alleles as the human, chimpanzee, and gorilla versions of the MsH42 region, respectively. The homogeneity of the sequence data sets was evaluated with the incongruence length difference test. The model that best fits the standard tree primate topology to the sequences was selected by the likelihood ratio test available in Modeltest version 3.0. Phylogenetic analysis are based on the neighbor-joining (distances were estimated under maximum-likelihood criteria), maximum-parsimony, and maximum-likelihood (GTR+G model) methods available in the PAUP* version 4 program (Swofford 1998). The node support was evaluated by bootstrapping using 1,000 resampled versions of the original data sets. All these analytical methods are reviewed by Thornton and DeSalle (2000). The likelihood values for the standard primate topology and the observed topologies were compared using the parametric Kishino-Hasegawa test (Kishino and Hasegawa 1989).
The GenBank accession numbers for the DNA sequences in this study are AY270192 to AY270201 for the minisatellite region, AY268962 to AY268968 for exon 5, and AY268969 to AY268975 for exon 6.
Results and Discussion
To establish whether the MsH42 region is present in primates, we carried out several PCR experiments using different combinations of primers (fig. 1). Primers PS1/PS2 amplified the genomic DNA from chimpanzee, gorilla, orangutan, and gibbon. However, in the case of gorilla DNA, the first round of PCR produced weak amplifications, and reamplification was therefore carried out on agarose-purified products. Attempts to amplify the genomic DNA of Old World monkeys (macaques and mandrill), New World monkeys (tamarin), and prosimian lemurids (Lemur, Eulemur, and Varecia) with those primers were unsuccessful. For this reason, we tried the amplification with primers P1/P2, which also span the minisatellite region. These primers gave successful amplification with the macaque and tamarin genomic DNA but failed to amplify the mandrill and lemurid DNAs, even when different combinations of the four primers were used. Mandrill genomic DNA was finally amplified with primers PC1/PC2, but once again the results with the prosimian DNA were negative. In table 1, the number of repeats, the presence of repeat variants, and the size of all primate MsH42-related alleles are summarized.
We performed amplifications with E5.1/PC2 to determine if the MsH42 region was also localized within an intron in primates. The results of these PCRs demonstrated the intronic localization of the MsH42-related minisatellite in primates (data not shown). On the other hand, we sought for the presence of the MsH42 region in the mouse, rat, zebrafish, and Drosophila genomic databases (www.ensembl.org). The result of these searches did not reveal any homologous sequences to the MsH42 region in those genomes. Because the genomes of both murine species contain the exons that flank MsH42, and there is no trace of the minisatellite, it is reasonable to conclude that the MsH42 minisatellite was originated during primate evolution.
To analyze the organization of the MsH42 region in the primates, the PCR products were sequenced. Figure 2 shows the alignment of a representative allele from each species in relation to the human MsH42 short allele. The tamarin version of the MsH42 locus has two repeat variants (A and C) common to humans, intercalated with other similar repeats, reminiscent of a pre-MsH42 region. In the Cercopithecoidea monkeys, macaque and mandrill, the minisatellite contains seven repeat units with the same arrangement in both species except for a G/A transition; this result agrees with the close phylogenetic relationship between these monkeys. Furthermore, in these primates, a constant characteristic of MsH42 emerges for the first time in all species: A is the first variant repeat and C1 is the last one (from 5' to 3'). In the gibbon DNA, the organization of the minisatellite is rather unusual and consists of 16 repeats, four of them restricted to this species (A5, C3, C4, D). The orangutan minisatellite has a repeat composition that is very similar to humans, and among its 13 repeats, only B4 is unique to this species. The panorama in the African great apes changes dramatically and the organization of MsH42 becomes highly homologous to the human locus. Thus, the analysis of gorilla samples revealed the existence of three alleles (table 1) with only one variant repeat (C2) absent in the human alleles. All gorilla alleles share a group of nine repeats at the 5' end of the minisatellite and another group of five repeats at the 3' end. Consequently, the differences among these alleles lie in the central portion of the minisatellite. The chimpanzees alleles (table 1) show an organization very similar to the human short allele, with the exception of two rare repeats (C2 and B3). It is worth noting that the GOR1 and CHI1 alleles are highly homologous to the human short allele, pointing towards their common origin from an ancestral array.
FIG. 2. Alignment among primate sequences of MsH42 region (5' flanking, 3' flanking, and minisatellite MsH42), exon 5, and exon 6. The human sequence is represented by the short allele, and chimpanzee and gorilla are represented by the CHI1 and GOR1 alleles, respectively. In the other primates, the only known allele is shown. To facilitate allele comparisons, the alignment of the minisatellite is shown using the one letter code (see table 1 for the key code). 5'HCS and 3'HCS boxes indicate the conserved sequences that include the PC1 and PC2 primers (underlined) and the 5f1 and 3f1 DNaseI footprints. chi boxes show the conservation of chi (GCTGGTGG) prokaryotic recombination hotspot. In the aligned sequences of exon 5, lined boxes mark two codons coding for methionine in almost all nonhuman primates and for threonine and valine in humans
Several conclusions can be drawn from the comparisons among primate alleles: (1) our findings agree with the belief that the African apes are more closely related to humans than the orangutans in Asia (Miyamoto, Slightom, and Goodman 1987; P??bo 2003); (2) the repeats A4 and B1 are exclusive of humans; (3) the repeats A and C probably are the most ancient; (4) the repeats A and C1 mark the 5' and 3' ends of the minisatellite, respectively; and (5) the existence of repeat blocks conserved in almost all species suggests that their presence could be necessary to construct this minisatellite.
Genomic sequence comparisons among distant species are a useful tool to identify regulatory elements present in the noncoding fraction of the genome. For example, sequence-specific conservation of noncoding DNA may imply functional constraint on these sequences and slower rates of molecular evolution (Ludwig 2002). Close examination of the sequences flanking MsH42 revealed an important conservation among primates (fig. 2). In particular, there are two highly conserved stretches (HCS) at both sides of the minisatellite, denoted as 5'HCS and 3'HCS. The 5'HCS comprises 29 bp with only two nucleotide changes in the tamarin, whereas the 3'HCS comprises 26 bp and shows only two transitions (A/G) in the macaque. It is likely that mandrill has both HCS because its genomic DNA was amplified with primers PC1/PC2, whose sequence is inside such regions. Phylogenetic footprinting (Tagle et al. 1988; Gumucio et al. 1992) and phylogenetic shadowing (Boffelli et al. 2003) have been used to identify putative regulatory elements, exploiting alignments across numerous distantly related or closely related species. We have previously detected protein-DNA interactions in this area by band-shifting experiments (Boán et al. 1997). Therefore, we carried out footprinting analysis to determine if these HCS were specifically recognized by proteins, thereby reflecting a function within the genome. For this purpose, we used the probes Fp1 and Fp2 (fig. 1). Fp1 contains the first 127 bp of the MsH42 region, including the 5'HCS and the first nine repeats. Fp2 has the last nine repeats of MsH42 and the next 175 bp downstream MsH42, including the 3'HCS. The results obtained with both orientations (upper and lower strands) of Fp1, using nuclear extracts from liver and brain, demonstrated the presence of two protected zones, 5f1 and 5f2, localized in the 5' flanking region of MsH42 (fig. 3). In the experiments performed with the Fp2 probe, we found two protections, 3f1 and 3f2, with the liver extract (fig. 3). Noteworthy, the 5f1 and 3f1 footprints include the 5'HCS and the 3'HCS, respectively, demonstrating the existence of a protein(s) that specifically interact with these conserved sequences, resembling phylogenetic footprints. Testes nuclear extract did not show any protected zone (fig. 3), indicating that testes may not express the protein(s) that recognizes such conserved zones. Taking into account that the MsH42 locus is situated within an intron, it is tempting to speculate that these HCS could play a role as regulatory elements of the Q9ULM1 gene. It should be mentioned that the footprinting experiments were done using rat nuclear extracts. Given that the rat genome does not contain the MsH42 region, it is reasonable to conclude that the sequences involved in the generation of footprints are present elsewhere in the rat genome and, hence, highly conserved.
FIG. 3. DNaseI footprinting analysis with the Fp1 and Fp2 probes. Upper strand corresponds to the sequence shown in figure 2, whereas lower strand is the complementary one. 5f1, 5f2, 3f1, and 3f2 indicate the DNaseI protected zones whose sequences are shown at the bottom. Arrows mark the start of MsH42 minisatellite. Lanes: G+A indicates Maxam and Gilbert G+A sequencing reactions; Control indicates reactions with BSA; Liver, Brain, Testes, indicates reactions with the corresponding nuclear extract
To compare the evolutionary dynamics of the intronic MsH42 region with respect to the proximal coding sequences, we amplified and sequenced the MsH42 neighbor exons 5 and 6 (fig. 1). The exon 5 amplified in all the primates studied, except in the lemurids, and the majority of changes were synonymous at the third codon position (fig. 2). We noticed that there is a methionine codon in all nonhuman primates that codes for threonine in humans, suppressing in this way a putative initiation codon. Perhaps this potential initiation codon might serve to produce more than one protein from the same mRNA by a mechanism involving translational initiation using internal ribosome binding sites (Oh and Sarnow 1993). Exon 6 is highly conserved in all primates except tamarin, all the changes being synonymous with respect to the human exon (fig. 2).
A phylogenetic analysis using the DNA sequences from the minisatellite region and exons 5 and 6 was performed. The results from the incongruence-length difference test indicate a significant character congruence (P = 1) that allows us to reconstruct a neighbor-joining topology from the total data set (data not shown). The phylogenetic tree so obtained is identical to the standard hypothesis about the phylogenetic relationships of primates (Miyamoto, Slightom, and Goodman 1987). Under the parsimony criterion, the total and the minisatellite flanking data sets also reflect the widely accepted primate tree, whereas alternative topologies were recovered from the exons and minisatellite sequences (data not shown). However, the consistency-index values as well as the Kishino-Hasegawa test indicate that these sequences make a consistent statement with respect to the standard primate topology. These results indicate that the mutational events observed in the minisatellite MsH42 are in agreement with the primate evolutionary history described for other noncoding sequences (Saitou and Ueda 1994; Apoil and Blancher 2000). Moreover, the estimation of parameter ( = 0.256) from population data suggests a mutation rate in the range of those estimated for other intron sequences in human genome (Huang et al. 1998).
The results presented here provide strong evidence that MsH42 was originated within an intron from a progenitor sequence, without a well-defined minisatellite structure, that experienced mutations leading to the formation of the first MsH42 repeat variants (fig. 4). Such progenitor sequence could have been originated by slipped-strand mispairing and unequal crossing-over between noncontiguous repeats formed by chance (Levinson and Gutman 1987; Haber and Louis 1998; Taylor and Breden 2000). The existence of the MsH42-like region in tamarin indicates its ancient generation during the earliest primate lineage evolution, before the divergence between Old World and New World monkeys about 40 MYA (Goodman 1999). This belief is strongly supported by the lack of amplifications in three prosimian species and because MsH42 homologous sequences could not be retrieved from the genomic databases of other organisms.
FIG. 4. Hypothetical evolutionary history of the minisatellite MsH42. The minisatellite organization of the different alleles for each species is shown. The repeat units are represented by a letter code (see table 1). It should be noted that the code was modified with respect to the previous one (Boán et al. 1998, 2002). Repeats present in human MsH42 are colored, and its probable appearance in the primate lineage is shown at the internodes along the tree. R in the common ancestor array indicates repeats of unknown sequence. Divergence times were taken from P??bo (2003) and Goodman (1999). Mys indicates million years ago
The homology between the human MsH42 and the corresponding locus in the African great apes, chimpanzee and gorilla, is remarkable. Comparison of the human short allele with the CHI1 and GOR1 alleles revealed an almost complete identity among the three sequences. There are two blocks of repeats, A-B-A-C-A1-B-A2-C-A and A-A-C-A-B2-A2-C-A-C-A-A3, identical in the three species, suggesting the existence of a common ancestral allelic array that has not become extinct in the actual human, chimpanzee, and gorilla populations. It is possible that the present polymorphism in humans and African great apes was originated from these highly homologous alleles. Surprisingly, the allele GOR1 is more similar to the human and chimpanzee alleles than to the other gorilla alleles (GOR2 and GOR3). This finding raises the possibility that several allelic forms of the MsH42 region coexisted in the ancestral genetic pool and that GOR2 and GOR3 alleles evolved from any of those forms, whereas GOR1, CHI1, and the HUM(S) alleles evolved from another allelic array. Nonetheless, the existence of these highly homologous sequences in the three species indicates that this ancestral common array should have been the most abundant allele. In fact, the identification of shared alleles between species rather than the relative frequencies is an accurate method to establish ancestral/derived status of polymorphic alleles (Iyengar et al. 1998).
The theory that modern humans originated in Africa is strongly supported by analysis of genetic variation in people today and from fossil discoveries (Stinger 2003). The existence of a highly homologous MsH42 allele in humans and African great apes, together with the fact that the frequency of the short allele in African humans (0.48 ± 0.07) is three times higher than the observed frequency in Europeans (0.16 ± 0.02) (Boán et al. 2002), supports the "Out of Africa" hypothesis. What is the origin of human MsH42 polymorphism? According to our results, the sequence of events that generated the present MsH42 polymorphism could have been as follows: First, there was the short allele coming from an evolutionary ancestor common to the great apes, second, a single duplication of a repeat block produced the long allele, third, the long allele generated the middle one by a deletion (fig. 4). The maintenance of the repeat arrangement in the three alleles points towards the mispairing of repeat blocks as the most probable mutational events in the human MsH42, as it has been proposed for many minisatellites (Charlesworth, Sniegowski, and Stephan 1994).
A useful approach to study human evolution at the molecular level is to consider our genome as a mosaic in which each DNA segment has its own evolutionary history (P??bo 2003). Our results allowed us to figure out when the human MsH42 began to exist early in primates, the birth of the minisatellite, and how this locus evolved to become MsH42 in humans. We believe that the present work is a worthwhile contribution to the knowledge of the low-variability minisatellites origin and that the evolutionary analysis of the MsH42 region is a small step toward the understanding of how the human genome has been made.
Acknowledgements
We thank M. Delclaux and coworkers from the Zoo of Madrid for providing us with the primate blood samples. We also thank J. Bertranpetit, A. Blancher, J. Bullerdiek, S. P??bo, P. Rogalla, W. Schempp, and R. Wimmer for giving us primate genomic DNA samples. Finally, we thank J. L. Caeiro for the African DNA samples. We would like to acknowledge J. M. Rodríguez for his help in some experiments and P. Barros for her help with figures. This project was funded by a grant of the Spanish Ministerio de Ciencia y Tecnología (BMC2001-3242). M.G.B. is recipient of a FPU grant from the Spanish government.
Literature Cited
Apoil, P.-A., and A. Blancher. 2000. Rh gene evolution in primates: study of intron sequences. Mol. Biol. Evol. 17:127-136.
Boán, F., A. González, J. M. Rodríguez, and J. Gómez-Márquez. 1997. Molecular characterization of a new human minisatellite that is able to form single-stranded loops in vitro and is recognized by nuclear proteins. FEBS Lett. 418:251-257.
Boán, F., J. M. Rodríguez, and J. Gómez-Márquez. 1998. A non-hypervariable human minisatellite strongly stimulates in vitro intramolecular homologous recombination. J. Mol. Biol. 278:499-505.
Boán, F., J. M. Rodríguez, S. Mouri?o, M. G. Blanco, A. Vi?as, L. Sánchez, and J. Gómez-Márquez. 2002. Recombination analysis of the human minisatellite MsH42 suggests the existence of two distinct pathways for initiation and resolution of recombination at MsH42 in rat testes nuclear extracts. Biochemistry 41:2166-2176.
Boffelli, D., J. McAuliffe, D. Ovcharenko, K. D. Lewis, I. Ovcharenko, L. Pachter, and E. M. Rubin. 2003. Phylogenetic shadowing of primate sequences to find functional regions of the human genome. Science 299:1391-1394.
Charlesworth, B., P. Sniegowski, and W. Stephan. 1994. The evolutionary dynamics of repetitive DNA in eukaryotes. Nature 371:215-220.
Goodman, M. 1999. The genomic record of humankind's evolutionary roots. Am. J. Hum. Genet. 64:31-39.
Gray, I. C., and J. A. Jeffreys. 1991. Evolutionary transience of hypervariable minisatellites in man and the primates. Proc. R. Soc. Lond. B Biol. Sci. 243:241-253.
Gumucio, D. L., H. Heilstedt-Williamson, T. A. Gray, S. A. Tarlé, D. A. Shelton, D. A. Tagle, J. L. Slightom, L. Goodman, and S. F. Collins. 1992. Phylogenetic footprinting reveals a nuclear protein which binds to silencer sequences in the human gamma and epsilon globin genes. Mol. Cell. Biol. 12:4919-4929.
Haber, J. E., and E. J. Louis. 1998. Minisatellite origins in yeast and humans. Genomics 48:132-135.
Huang, W., Y. X. Fu, B. H. Chang, X. Gu, L. B. Jorde, and W. H. Li. 1998. Sequence variation in ZFX introns in human populations. Mol. Biol. Evol. 15:138-142.
Iyengar, S., M. Marinero, C. Deinard, H. C. Rosenbaum, G. Sirugo, C. Castiglione, J. R. Kidd, and K. K. Kidd. 1998. Analyses of cross species polymerase chain reaction products to infer the ancestral state of human polymorphisms. DNA Seq. 8:317-327.
Jeffreys, A. J. 1987. Highly variable minisatellites and DNA fingerprints. Biochem. Soc. Trans. 15:309-317.
Kishino, H., and M. Hasegawa. 1989. Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data and the branching order in hominoidea. J. Mol. Evol. 29:170-179.
Levinson, G., and G. A. Gutman. 1987. Slipped-strand mispairing: a major mechanism for DNA sequence evolution. Mol. Biol. Evol. 4:203-221.
Ludwig, M. Z. 2002. Functional evolution of noncoding DNA. Curr. Opin. Genet. Dev. 12:634-639.
Miyamoto, M. M., J. L. Slightom., and M. Goodman. 1987. Phylogenetic relations of humans and African apes from DNA sequences in the -globin region. Science 238:369-373.
Oh, S., and P. Sarnow. 1993. Gene regulation: translational initiation by internal ribosomal binding. Curr. Opin. Genet. Dev. 3:295-300.
Olsen, K. M. 1999. Minisatellite variation in a single-copy nuclear gene: phylogenetic assessment of repeat length homoplasy and mutational mechanism. Mol. Biol. Evol. 16:1406-1409.
P??bo, S. 2003. The mosaic that is our genome. Nature 421:409-412.
Saitou, N., and S. Ueda. 1994. Evolutionary rates of insertion and deletion in noncoding nucleotide sequences of primates. Mol. Biol. Evol. 11:504-512.
Stinger, C. 2003. Human evolution: out of Ethiopia. Nature 423:692-694.
Swofford, D. L. 1998. PAUP*: phylogenetic analysis using parsimony (*and other methods). Version 4. Sinauer Associates, Sunderland, Mass.
Tagle, D. A., B. F. Koop, M. Goodman, J. L. Slightom, D. Hess, and R. T. Jones. 1988. Embryonic epsilon and gamma globin genes of a prosimian primate (Galago crassicaudatus): nucleotide and aminoacid sequences, developmental regulation and phylogenetic footprints. J. Mol. Biol. 203:439-455.
Taylor, J. S., and F. Breden. 2000. Slipped-strand mispairing at non-contiguous repeats in Poecilia reticulata: a model for minisatellite birth. Genetics 155:1313-1320.
Thornton, J. W., and R. DeSalle. 2000. Gene family evolution and homology: genomics meets phylogenetics. Annu. Rev. Genomics Hum. Genet. 1:41-73.
Vergnaud, G., and F. Denoeud. 2000. Minisatellites: mutability and genome architecture. Genome Res. 10:899-907.(Francisco Boán, Miguel G.)