Molecular Biology and Evolution

MBE Advance Access originally published online on December 29, 2005
Molecular Biology and Evolution 2006 23(4):798-806; doi:10.1093/molbev/msj088

This Article Abstract FREE Full Text (PDF) Supplementary Material All Versions of this Article: 23/4/798    most recent msj088v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (2) Request Permissions Google Scholar Articles by Iwashita, S. Articles by Osada, N. Search for Related Content PubMed PubMed Citation Articles by Iwashita, S. Articles by Osada, N. Social Bookmarking          What's this?

© The Author 2005. 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

Research Article

A Tandem Gene Duplication Followed by Recruitment of a Retrotransposon Created the Paralogous Bucentaur Gene (bcnt^p97) in the Ancestral Ruminant

Shintaro Iwashita^*,, Sadao Ueno^*,, Kentaro Nakashima^*,, Si-Young Song^*, Kenshiro Ohshima, Kazuaki Tanaka, Hideki Endo^||, Junpei Kimura^¶, Masamichi Kurohmaru^#, Katsuhiro Fukuta^**, Lior David and Naoki Osada

^* Mitsubishi Kagaku Institute of Life Sciences (MITILS), Tokyo, Japan; Department of Environment and Natural Sciences, Yokohama National University, Kanagawa, Japan; Hitachi Instruments Service Co., Ltd., Tokyo, Japan; Department of Animal Science and Biotechnology, Azabu University, Kanagawa, Japan; ^|| National Science Museum, Tokyo, Tokyo, Japan; ^¶ Department of Veterinary Medicine, Nihon University, Kanagawa, Japan; ^# Department of Veterinary Anatomy, The University of Tokyo, Tokyo, Japan; ^** Laboratory of Animal Morphology and Function, Nagoya University, Nagoya, Japan; Department of Biochemistry, Stanford University; and National Institute of Biomedical Innovation, Osaka, Japan

E-mail: siwast{at}libra.ls.m-kagaku.co.jp.

	Abstract

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 
Retrotransposable element-1 (RTE-1) is a class of long interspersed nucleotide elements that contain in its open reading frame an apurinic/apyrimidinic endonuclease domain (AP-END) and a reverse transcriptase domain. Ruminants have a clade-specific RTE-1 (BovB/RTE). The bovine bcnt gene (bucentaur or craniofacial developmental protein 1) has a duplicated paralog (bcnt^p97) in tandem that recruited an AP-END of BovB/RTE as a coding exon (RTE exon). We obtained sequence of the bcnt region from several animals and showed that other ruminants also have the bcnt^p97 with a conserved RTE exon while camels and pigs do not. Genomic Southern analysis showed that camels and pigs have multiple bcnt-related sequences but not BovB/RTE which bovines and lesser mouse deer have abundantly. These results indicate that the bcnt gene duplication followed by the creation of bcnt^p97 including recruitment of the RTE exon occurred in the ancestral ruminant about 55 MYA. The indication of time frame is supported by a phylogenetic analysis. Taken together with a result of differential tissue expression of the two bcnt paralogs, we conclude that bcnt^p97 was created concurrently with the early radiation of BovB/RTE in an ancestral ruminant and then acquired a novel function.

Key Words: gene duplication • ruminant • exonization • AP-endonuclease domain • reverse transcriptase domain • adaptive evolution

	Introduction

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 
Accumulating genomic data for various organisms have revealed that transposable elements (TEs), such as short interspersed repetitive elements (SINEs) and long interspersed repetitive elements (LINEs), may work as a direct source or a regulator for novel gene creation (Makalowski 2000; Long 2001; Nekrutenko and Li 2001; Lev-Maor et al. 2003; Brosius 2005). Retrotransposable element-1 (RTE-1) is one of the shortest classes of LINEs and comprises an apurinic/apyrimidinic endonuclease domain (AP-END) and a reverse transcriptase domain (RTD) (Malik and Eickbush 1998). In bovines, RTE-1 was initially reported as a bovine Alu-like dimer driven family, which contains both AP-END and RTD accompanied by a SINE cassette (Szemraj et al. 1995). Although RTE-1 elements are widely distributed in eukaryotes, bovine RTE-1 (BovB/RTE) has been found only in ruminants, suggesting that it has been transmitted horizontally from Squamata (Kordis and Gubensek 1998; Zupunski, Gubensek, and Kordis 2001).

We have identified a ruminant-specific protein, p97Bcnt, which has a 325-amino acid (aa) region derived from an AP-END of BovB/RTE (Nobukuni et al. 1997; Takahashi et al. 1998). In bovines, bcnt^p97 is located 6 kb upstream of a bcnt in a tandem manner, and both genes express proteins. Bcnt protein lacks the RTE-1–derived region but has a highly conserved 82-aa region at the C-terminus that is not present in p97Bcnt. Furthermore, while p97Bcnt contains two intramolecular repeat (IR) units at the C-terminus, ancestral Bcnt contains only one IR unit in the middle of the molecule (fig. 1; Iwashita et al. 2003). These results provided a scenario in which a BovB/RTE directly contributed to the creation of a novel gene, bcnt^p97, by its insertion into one of the ancestral duplicated genes (Makalowski 2003).

View larger version (29K): [in this window] [in a new window]   FIG. 1.— Bcnt and p97Bcnt of LMD and bovines and their different tissue expression. (A) A schematic comparison of domain structure of Bcnt and p97Bcnt between bovines and LMD. On top is the structure of a BovB/RTE element in which light and dark gray boxes indicate an AP-END and a RTD, respectively. The square at the left of RTE-1 indicates the ambiguity of its 5' region (Malik and Eickbush 1998). White boxes represent regions that are derived from RTE-1 but not AP-ENDs. The light gray boxes in the middle of p97Bcnt show the region derived from an AP-END of a BovB/RTE. The N-terminal meshed boxes show the acidic regions, and the horizontal striped boxes represent two IR units. The black box at the C-terminus of Bcnt indicates a highly conserved region that was not found in p97Bcnt. The numbers above boxes indicate total amino acid residues. (B) Differential tissue expression of Bcnt and p97Bcnt. Protein extract from brain, lung, heart, and liver of bovines were subjected to immunoblotting either with anti-Bcnt C-terminal peptide antibody (upper panel) or with anti-bovine p97Bcnt monoclonal antibody mixtures (lower panel).

 
Gene duplication is one of the major ways that initiate the creation of new gene structures and forms a more refined signaling network in higher organisms. But due to the cumulative nature of evolutionary changes, it is generally difficult to track at the molecular level what the evolutionary processes that create two diverged paralogs are (Lynch and Katju 2004). Because the gene bcnt^p97 might have arisen by a clear combination of gene duplication and insertion of BovB/RTE, the evolutionary study of bcnt and bcnt^p97 may provide a good opportunity to understand the process of novel gene creation in higher organisms.

The genomic organization of bcnt and bcnt^p97 was previously described only in bovines (Iwashita et al. 2003). Very limited amount of sequence information is available from other ruminants, which makes it difficult to study the time course of the gene duplication and the creation of bcnt^p97. In this study, we obtained sequence information from several species including pigs, camels, and lesser mouse deer (LMD), which represents the earliest divergent taxa of ruminants (Hassanin and Douzery 2003). Then we performed a comparative analysis of the genomic organization of bcnt region and determined the time frame and evolutionary patterns of the bcnt paralogs.

	Materials and Methods

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 
DNA and RNA Samples
LMD (Tragulus javanicus) samples were collected from a livestock farm in Kota Kinabalu, Sabah, Malaysia, and their muscles and livers were immediately preserved. Genomic DNA of greater mouse deer (GMD, Tragulus napu, Endo et al. 2002) and sika deer (Yakushika deer, Cervus nippon yakushikae; Iwashita et al. 1999) were also prepared. Camel (Camelus dromedarius) samples were extracted from frozen spleen, a gift from Y. Kurosawa (Fujita Health University, Toyoake). Muntjac (Muntiacus reevesi, female) was captured in Chiba prefecture, Japan, and the muscle was immediately stocked in RNAlater (Ambion, Austin, Tex.). Pig (Sus scrofa) DNA was a gift from Y. Takagaki (Kitasato University School of Medicine, Sagamihara). Human placenta DNA was obtained from Clontech (Palo Alto, Calif.).

Isolation of bcnt and bcnt^p97 cDNAs
Total RNA of aforementioned tissues from LMD, camels, and muntjacs were prepared using the RNeasy RNA preparation kit (Qiagen, Hilden, Germany). The bcnt cDNAs were obtained by reverse transcription–polymerase chain reaction (RT-PCR) according to the manufacturer's protocol (Invitrogen) and amplified with gene-specific primers (all primers are listed in Supplementary table S1, Supplementary Material online). The melting temperature (Tm) of each oligonucleotide was calculated by BioMath Calculators (www.promega.com/biomath/calc11.htm). PCR was performed with 1 µl of cDNA template, 2.5 mM MgCl₂, 200 µM of each deoxynucleoside triphosphate, 10 pmol of each primer, 1.25 units of LA Taq polymerase (Takara, Otsu, Japan) in a total volume of 50 µl. The PCR protocol included the following steps: denaturation at 94°C for 2 min; five amplification cycles of denaturation at 96°C for 20 s, annealing at Tm + 5°C for 30 s, and extension at 72°C for 1 min for the bcnt cDNA or 3 min for the bcnt^p97cDNA; then additional 25 cycles under the same conditions but with annealing temperature at Tm; and finally an elongation of 4 min at 72°C. PCR products were cloned into a pCR2.1-TOPO vector (Invitrogen, Carlsbad, Calif.) or sequenced directly. Sequence of pig bcnt cDNA was obtained from the Institute for Genomic Research Porcine Gene Index (SsGI), and its open reading frame (ORF) sequence was confirmed.

Construction of LMD Genomic Fosmid Library
The construction of LMD fosmid library and screening bcnt and bcnt^p97 is described in Supplementary Materials and Methods and figure S1 (Supplementary Material online).

Isolation of Pig BAC Clones
A pig genomic bacterial artificial chromosome (BAC) library (Suzuki et al. 2000) was screened using each set of primers (Supplementary table S1, Supplementary Material online) for bcnt and the processed bcnt (GenBank: AB088356). Two clones containing bcnt (711F3 and 990B3) and one clone with its processed gene (544F10) were isolated by Hiroshi Yasue (National Institute of Agrobiological Sciences, Tsukuba). All plasmids were purified using a purification kit (Qiagen), and their DNA sequences were determined by primer-walking method.

Preparation of Camel Adaptor-Ligated Genomic Library
Camel DNA was digested separately with DraI, EcoRV, PvuII, or StuI, and each digest was ligated with an adaptor according to the manufacturer's protocol of Genome Walker Universal kit (BD Biosciences, Palo Alto, Calif.).

Sequencing of bcnt^p97 from Other Ruminants
The intergenic regions between bcnt and bcnt^p97 of sika deer and GMD and the region of bcnt^p97 intron 4 from GMD were amplified by PCR from their genomic DNA using each set of primers (Supplementary table S1, Supplementary Material online). All the PCR products were subcloned into pCR2.1-TOPO vector and sequenced using Big Dye terminator mix and an ABI3100 sequencer (Applied Biosystems, Foster City, Calif.).

Southern Blotting Analyses
Southern blotting and hybridization with ³²P-labeled probes were performed as previously described (Hon-Nami et al. 2004) except that here we digested the genomic DNAs with EcoRI and used a different nylon filter (GeneScreen Plus, NEN, Boston, Mass.). To check the Southern sensitivity, 955-bp fragment of LMD RTE exon (Hon-Nami et al. 2004) was mixed with 10 µg of EcoRI digests from either camel or pig genomic DNA at molar ratios 0, 1, and 5, during the ethanol precipitation before separation in agarose. By random primer labeling (Amersham) with [-³²P] deoxycytidine triphosphate, four probes were prepared: (1) a pig 460-bp fragment consisting of 220 bp of bcnt exon 3 with its 5'and 3' intron regions (probe A), (2) the corresponding LMD bcnt^p97 exon 3 with 440 bp (probe B), (3) the LMD RTE exon (probe C, Hon-Nami et al. 2004), and (4) a fragment of 131 bp of human LOC124491 (probe D). The fragment of probe D was amplified from human placenta DNA by PCR (see Supplementary table S1, Supplementary Material online, for primers). The specific activity of labeled probes were 1–2 x 10⁹ dpm/µg. Hybridization with probe A or B was performed at 60°C for 14 h for genomic DNA or 2 h for BAC clones, and the filter was washed twice with 2 x standard saline citrate (SSC)/1% sodium dodecyl sulfate (SDS) at 57°C for 30 min, then with 0.5 x SSC/0.1% SDS at 50°C for 10 min. After analyzing the result, the filter was stripped in boiling 0.1 x SSC/1% SDS for 10 min x 3 and then confirmed for zero residual radioactivity before used for the next probe. Hybridization with probe C or D was performed as previously described (Hon-Nami et al. 2004).

Immunoblotting
LMD extract was prepared by slicing and mincing the frozen muscle on dry ice followed by homogenizing in a Dounce homogenizer as previously described (Iwashita et al. 2003). Bovine protein extract from several tissues was prepared as previously described (Nobukuni et al. 1997). Each 40 µg sample was separated on 10% SDS polyacrylamide gels and subjected to immunoblotting with anti-p97Bcnt monoclonal antibodies or anti-Bcnt C-terminal peptide antibody as previously described (Iwashita et al. 2003). For epitope sequences and antibody specificities, see Supplementary table S2 (Supplementary Material online).

Sequence Analysis
Sequences alignment was done using ClustalX (Thompson et al. 1997) or Blast 2 sequences (Tatusova and Madden 1999). TEs in the genomic sequences were analyzed using the program RepeatMasker (A. F. A. Smit and P. Green, RepeatMasker at http://www.repeatmasker.org/cgi-bin/WEBRepeatMasker).

Phylogenetic Analysis
We used various 5' sequences that encode the N-terminal region of both bcnt and bcnt^p97: human, mouse (Takahashi et al. 1998), bovine (Iwashita et al. 2003), LMD, muntjac, camel, and pig (this study). These sequences were aligned using ClustalX with default parameters (Thompson et al. 1997), and the multiple alignment was used to calculate a phylogenetic tree by the maximum likelihood method implemented in the PAML program package (Yang 1997).

	Results

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 
LMD have bcnt^p97 Similar to Bovines
We first amplified a partial cDNA of LMD bcnt by RT-PCR from liver RNA and determined its sequence. Then we screened a genomic library of LMD to determine the genomic sequence of this region from eight fosmid clones containing bcnt sequence. We found that the bcnt^p97 of LMD locates upstream of the bcnt locus in tandem, similar to the arrangement found in bovines. LMD bcnt and bcnt^p97 consists of seven and eight exons, respectively, and the LMD exon-intron structure is essentially the same as that of the bovine genes. The overall gene structure is presented in figure S1 (Supplementary Material online), and more specific features will be discussed later. The LMD Bcnt is 298 aa long and 1 aa longer than bovine Bcnt with 91% sequence identity (fig. 1A). On the other hand, LMD p97Bcnt is 574 aa long, and compared to bovines it is shorter by 6 aa upstream to the RTE exon and by 12 aa at its C-terminus (fig. 1A). We confirmed the protein size of both genes in LMD and bovines by immunoblotting (fig. S2, Supplementary Material online). In accordance with the difference in amino acid length, LMD p97Bcnt was slightly smaller than bovine p97Bcnt by SDS–polyacrylamide gel electrophoresis.

The Paralogs Show Differential Tissue Expression
The presence of p97Bcnt in both bovines and LMD suggests that bcnt^p97 may have acquired a different function from the ancestral Bcnt. To gain insight into this issue, we examined the expression of both proteins in bovine tissues by western blot analysis (fig. 1B). Bcnt is predominantly expressed in brain, a little in liver, but undetectable in lung and heart. On the other hand, p97Bcnt is strongly expressed in the brain and moderately in the lung and liver. Furthermore, while p97Bcnt significantly localizes in the nuclei as an AP-endonuclease does (Iwashita et al. 1999), Bcnt exclusively exists in the cytosol (Nakashima K, Iwashita S, Kato C, Sasaki M, Osada N, Song S-Y, in preparation). These results strongly suggest acquisition of a different function by the paralog.

Tandem bcnt Duplication and bcnt^p97 Creation Are Discrete to Ruminants
To estimate when the ancestral bcnt tandemly duplicated and the RTE exon was acquired, we first examined the copy number of bcnt in pigs (Suidae) and camels (Tylopoda, Camelidae), which are close relatives of ruminants. With two probes that cover 220 bp of the pig bcnt exon 3 or LMD bcnt exon 3 (fig. 2A), we found that both pigs and camels have multiple bcnt-related sequences similar to bovines and LMD. When using a probe of LMD RTE exon, we scarcely found signals in pigs and camels, while very strong signals were found in bovines and LMD (fig. 2B). Densitometric analysis of intensities in bovines, LMD, camels, and pigs showed a signal ratio of approximately 5,000:250:2:1, respectively. We thus conclude that expansion of BovB/RTE did not occur in both pigs and camels.

View larger version (59K): [in this window] [in a new window]   FIG. 2.— Multiple copies of bcnt-related genes and radiation of RTE exon sequences. (A) Genomic DNA from bovines, LMD, camels, and pigs were digested and subjected to Southern blot analysis with 32P-labeled probe covering 220 bp of a porcine bcnt exon 3 (left panel) or with a fragment covering the LMD bcntp97 exon 3 (right panel). (B) EcoRI digests of camel or pig DNA were mixed with a fragment of RTE exon from LMD at a molar ratio of 0, 1, or 5 (see ratio below lanes). Samples were subjected to Southern blot analysis with 32P-labeled LMD RTE exon as probe. The left part is a short time exposure that demonstrates high intensity of hybridization signals in bovines. After the image was taken, the bovine lane was cut out, and the membrane was exposed 10 times longer to obtain the image of the right part that demonstrates the difference between LMD, camels, and pigs. The arrow indicates the position of externally added fragment of RTE exon.

 
To clarify the pig bcnt gene organization, we used three BAC clones that included two allelic forms of bcnt (clones 711F3 with 120 kb in size and 990B3 with 130 kb, fig. 3A) and one processed bcnt pseudogene (544F10). Southern blotting of EcoRI or HindIII digests of these three clones with a probe covering porcine bcnt exon 3 showed a single band in all three BAC clones (fig. 3A, left panel), indicating that the BAC clones contain a single copy of bcnt. Next, we performed Southern analysis probing with a fragment of human LOC124491, which locates 13.5 kb upstream of the human bcnt on chromosome 16q23 (fig. 3A). We detected a positive signal in both clones 711F3 and 990B3 (fig. 3B, right panel). Taking into account the sizes of the two BAC clones, the relative location of bcnt in each BAC clone, and the genomic organization of human bcnt, the results imply that pig bcnt did not duplicate in a tandem fashion like the ruminant ancestral bcnt.

View larger version (37K): [in this window] [in a new window]   FIG. 3.— No tandem bcnt duplication in pigs. (A) Using the University of California–Santa Cruz Genome Bioinformatics Site, two pig BAC clones 711F3 and 990B3 (dashed boxes) that include bcnt were tentatively aligned to human BCNT (solid box) neighboring region on chromosome 16q23. The alignment was based on the determined sequences from both sides of the BAC clones (FW and RV) and of bcnt (Supplementary Materials and Methods and fig. S3, Supplementary Material online). Scale units on human chromosome are 50 kb relative to the first nucleotide of bcnt ORF, and exons 1–7 of human bcnt are presented by numbered vertical bars in the box. (B) Three pig BAC clones (544F10, 711F3, and 990B3) were digested and subjected to a Southern blotting with 32P-labeled probe covering 220 bp of a pig bcnt exon 3 (left panel) or with another probe corresponding to a fragment of human LOC124491 gene (right panel). FW, M13 forward; RV, M13 reverse.

 
In camels, we constructed its adaptor-ligated genomic library and determined the rough bcnt gene organization (Supplementary Materials and Methods and fig. S3, Supplementary Material online). Then we sequenced regions 5' of bcnt in both camels and pigs and found the hypothetical LOC388297 gene which is a processed gene and reported to be expressed in human teratocarcinoma (GenBank: AK056030) within 1 kb upstream of bcnt locus. Humans and dogs also have the LOC388297 gene about 1 kb upstream of bcnt locus (fig. 4). In bovines, the two similar sequences of LOC388297 are also located upstream of both bcnt and bcnt^p97 (fig. 4), implying that the segmental duplication harbored both bcnt and the LOC388297 gene. These results clearly show that the tandem duplication of bcnt that created bcnt^p97 occurred sometime after the Ruminantia-Suina-Tylopoda split before the Pecora-Tragulina divergence.

View larger version (11K): [in this window] [in a new window]   FIG. 4.— A schematic representation of bcnt region organization in several species. DNA sequences upstream of bcnt locus from humans, dogs, bovines, pigs, and camels were analyzed by Blast search and represented schematically. These sequences except of bovines are conserved and include the hypothetical LOC388297 gene within 1 kb upstream of bcnt locus (upper portion). In bovines, a highly fragmented LOC388297 (dashed boxes) was located upstream of both bcnt and bcntp97 (lower portion). All regions are flanked by a split MER82/MER2 DNA transposon shown by M.

 
Phylogeny Supports the Time Frame of bcnt Duplication for bcnt^p97 Creation
To estimate the time frame of bcnt^p97 creation in an alternative method, we constructed a phylogenic tree of the bcnt family based on the 5' part of the ORF from various mammals, namely humans, mice, bovines, LMD, muntjacs, camels, and pigs. Using a maximum likelihood method, we obtained a tree which implies the gene duplication happened after the Ruminantia-Tylopoda and before the Pecora-Tragulina divergences (time A in fig. 5). We then tested whether this tree topology has significantly higher probability than other topologies. The tree of figure 5 was compared with the other two hypotheses, gene duplication before the Ruminantia-Tylopoda (time B in fig. 5) and gene duplication before the Ruminantia-Suina divergence (time C in fig. 5), by the method of Kishino and Hasegawa (1989). The result significantly supported the gene duplication after the Ruminantia-Tylopoda divergence as shown in figure 5 (P = 0.003). We also performed an analogous test assuming a different phylogenic tree in which the Tylopoda diverged earlier than the Suina, proposed by analysis of SINE and LINE markers (Nikaido, Rooney, and Okada 1999). The result was consistent with the former test (P = 0.006). Thus, not only the evidence from the genomic organization but also the evidence from the nucleotide substitution pattern supports that the bcnt^p97 creation occurred after the Ruminantia-Tylopoda-Suina and before the Pecora-Tragulina divergence.

View larger version (8K): [in this window] [in a new window]   FIG. 5.— Phylogenetic analysis of Bcnt family. Phylogenetic trees of the Bcnt family were constructed based on the 5' parts of bcnt ORF, which encode the N-terminal regions of both Bcnt and p97Bcnt proteins. The arrow marked as A points out the divergence of Pecora and Tragulina within the ruminants lineage. Arrows marked by B and C point out the splits between Ruminantia-Tylopoda and Ruminantia-Suina, respectively.

 
The Sequence Arrangement of bcnt^p97-bcnt Locus Has Been Diversified in Ruminants
Divergence of Tragulina and Pecora was estimated to be around 50 MYA (Hassanin and Douzery 2003), and both taxa have p97Bcnt with a conserved RTE exon as shown above. Because BovB/RTE was potentially acquired by horizontal transfer into the ancestral ruminant (Zupunski, Gubensek, and Kordis 2001), it is of interest to compare the more subtle differences in the distribution of these elements at this bcnt^p97-bcnt locus between LMD and bovine lineages. We examined the distribution of BovB/RTE in the genomic regions of the bovines and LMD by the RepeatMasker program and found a few qualitative differences (fig. 6). One BovB/RTE is present in the intron 4 of only LMD bcnt^p97 (flanking the 5'of the RTE exon). Another element appears in the bcnt^p97-bcnt intergenic region of bovines, while that region in LMD contains a sequence of a processed matrin type 2 (zmat2) gene with a zinc finger motif (fig. 6). We also determined the intergenic regions of both GMD and sika deer and found that GMD have the processed zmat2 sequence similar to LMD and that sika deer have a BovB/RTE fragment like bovines (data not shown). Furthermore LMD have 30 Tragulidae-specific SINE-like repeats in this bcnt^p97-bcnt locus (GenBank: AB192410). These results show a dynamic sequence rearrangement that is associated with TEs and occurred after the Pecora-Tragulina divergence.

View larger version (14K): [in this window] [in a new window]   FIG. 6.— A schematic comparison of bcnt-bcntp97 locus between LMD and bovines. Exons are presented by numbered vertical bars, and scale units are kilobase relative to the first nucleotide of bcnt ORF. The closed arrowheads show the locations and orientations of BovB/RTE, and open arrowheads represent LINE/L1. An enlarged scheme of the region between bcntp97 exon 4 and bcnt exon 1 is shown in the upper and lower panels for bovines and LMD, respectively. Exon 5 of the bcntp97 gene is an RTE exon (light gray boxes). LMD have a larger intron 4 (7.6 kb) and a larger intron 5 (1.0 kb). Meshed boxes in intron 4 represent similar sequences between LMD and bovines. The 1.6-kb region 5' to RTE exon in LMD is similar to three fragments of BovB/RTE (in a reverse orientation, gray). The intergenic regions are highly similar except that bovines include a BovB/RTE fragment, while LMD include a highly homologous sequence of the processed matrin type 2 gene (zmat2), encoding a hypothetical zinc finger protein. A fragmented LOC388297 is adjacent to one side of zmat2 in bovines or to both sides in LMD as denoted by dashed boxes.

 

	Discussion

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 
In this study, we present strong evidence that the bcnt gene duplication involved in the creation of bcnt^p97 was confined to this genomic region and restricted to ruminants. The creation of bcnt^p97 involved two steps, namely tandem duplication and an RTE exon acquisition, which have probably occurred in the ancestral ruminant within a relatively short time period and before further diversification of ruminant species. In addition, we show evidence for a massive expansion of a ruminant-specific BovB/RTE that mostly occurred in the ancestral ruminant. Therefore, the acquisition of an RTE exon in the process of bcnt^p97 creation was very likely a part of early expansion of BovB/RTE.

Ruminants are a suborder suitable to study macroevolution: questions such as how mammals adapt to dramatic environmental change such as climate and vegetation conditions after the extinction at the Cetaceous-Paleogene era. Ruminants currently include more than 190 species with established phylogeny (Hassanin and Douzery 2003) that share specific physiological features such as a unique digestive system. Studies on the digestive enzymes of ruminants such as ribonuclease (Breukelman et al. 2001) and lysozyme (Wen and Irwin 1999) identified multiple copies of these genes and revealed the evolutionary relationship among them. To understand the mechanism of macroevolution of this suborder from a molecular perspective, changes mediated by retrotransposon radiation might be one of the key factors (Brosius 2005). Therefore, it is noteworthy that the BovB/RTE has probably been transferred horizontally (Kordis and Gubensek 1998; Zupunski, Gubensek, and Kordis 2001).

It is now clear by genome-wide studies that exonization of retrotransposons is not unusual and that unequal homologous recombination between LINEs sometimes occurred. DNA repetitive sequences may contribute to the creation of gene diversity much more than we have realized. The time frame of BovB/RTE expansion is one of the most critical questions in genome evolution of ruminants. Bovine interspersed repetitive elements such as Pst family are widely spread in ruminants but not in other Artiodactyla (Modi, Gallagher, and Womack 1996), which we also confirmed by Southern analysis probing with RTE exon from LMD bcnt^p97. Camels and pigs have very low hybridization signal for LMD RTE exon–like sequences. It is also noteworthy that distinct difference in amounts of BovB/RTE was observed between bovines and LMD (20-fold difference) even using the LMD RTE exon as a probe but not the bovine RTE exon (identity is 90.6%, fig. 2B). The result implies that BovB/RTE had further expanded differentially during diversification of ruminant species. In contrast to the BovB/RTE distribution, we show that camels and pigs have multiple copies of bcnt-related genes in nonorthologous genomic regions similar to bovines and LMD. Processed bcnt genes are already known in bovines (on chromosome 26, Iwashita et al. 2003 and also see Note Added in Proof), in pigs (Takahashi et al. 1998 and this paper), and in rats (on chromosome 6). Furthermore, in mouse the orthologous bcnt maps on chromosome 8, and the bcnt pseudogene was found on chromosome 13 (using the University of California–Santa Cruz Genome Bioinformatics Site; http://genome.ucsc.edu/). On the other hand, we previously reported a LMD 1.2-kb fragment containing an RTE exon as a processed bcnt^p97 (GenBank: AB005651, Takahashi et al. 1998). But we failed to confirm it in this study, and so we assume it was an error and replaced it with AB191483 in GenBank. Taken together, these results strongly suggest that the paralogous bcnt^p97 was created in the ancestral ruminant as an early event that coincides with the early RTE-1 expansion before the diversification of ruminants about 45–65 MYA (fig. 5).

While the RTE exon of bcnt^p97 was well conserved in all six ruminants, the bcnt^p97-bcnt locus had rearrangements of repetitive sequences, including Tragulidae-specific SINE-like elements in the LMD. In the intron 5' to the RTE exon of bcnt^p97, Tragulus (LMD and GMD) contain a BovB/RTE fragment, while bovines and sika deer have just an 80-bp fragment of it, and Giraffidae has an additional family-specific interspersed repeat with a microsatellite at that position (Hon-Nami et al. 2004). While camels, pigs, humans, and dogs have LOC388297 but no bcnt^p97 in the region upstream of bcnt, bovines contain a homologous sequence similar to LOC388297 upstream to both bcnt^p97 and bcnt (fig. 4). The presence of a sequence similiar to LOC388297 upstream to both paralogs suggests that this was the gene order prior to the duplication of bcnt. In ruminants, bovines and sika deer further contain a BovB/RTE fragment, whereas Tragulus include a sequence of the processed zmat2 gene with a cysteine-rich motif. (fig. 6). This motif is associated with most non-LTR retrotransposons in either their ORF1 or ORF2 C-terminal regions (Poulter, Butler, and Ormandy 1999). As a result, the bcnt^p97-bcnt intergenic region contains LOC388297 sequences that are associated with a fragment of RTE-1 in bovines and sika deer or mixed with the processed zmat2 gene sequence in Tragulus (fig. 6). All these characteristic regions are flanked with a split MER2 DNA element. These data clearly show that the bcnt duplication was a discrete event confined to this region of the genome and that these sequence rearrangements in the bcnt^p97-bcnt locus were accompanied by insertion of additional TEs during the evolution of ruminants. The overall differences in organization of the bcnt^p97-bcnt locus match the known phylogeny of the ruminants (Hassanin and Douzery 2003).

Conservation of a full AP-END of BovB/RTE in p97Bcnt of ruminants implies a selective advantage in establishment of bcnt^p97as a new gene. AP-endonuclease is a multifunctional enzyme with a role in suppression of cell stress, directly by repairing DNA damage and indirectly by regulating the redox state of various proteins that modulate transcription factors (Tell et al. 2005). Although we have no direct evidence so far that p97Bcnt has any of these activities, it maintains some characteristics common to AP-endonuclease. First, the RTE exon conserves the major active sites of AP-endonuclease such as metal-binding sites. A three-dimensional structure of RTE exon could be constructed by modeling using 1HD7 as a template of AP-endonuclease in the presence of AP-DNA and metal (Beernink et al. 2001). Secondly, p97Bcnt can be phosphorylated by casein kinase II (Iwashita et al. 1998) similar to AP-endonuclease (Fritz and Kaina 1999), and thirdly, p97Bcnt significantly localizes in nuclei suggesting its function as nuclear protein, while Bcnt localizes mostly in the cytosol (Iwashita et al. 1998; Nakashima K, Iwashita S, Kato C, Sasaki M, Osada N, Song S-Y, in preparation). Therefore, it is possible that an ancestral ruminant with a tandem duplication of bcnt recruited an AP-END of a BovB/RTE and suppressed the cellular stress caused by a dramatic environmental change, resulting in the creation of p97Bcnt. This perspective is consistent with the fact that various retrotransposons including human LINEs can be induced by environmental stimuli, such as UV light and heat shock (Liu et al. 1995; Morales, Snow, and Murnane 2003). On the other hand, bcnt is an evolutionary conserved gene, and Drosophila Bcnt (YETI) has a well-conserved C-terminal region and binds to a microtubule-based motor kinesin-I (Wisniewski, Tanzi, and Gindhart 2003). However, the binding was mediated by YETI N-terminus that shows low similarity to mammalian Bcnt. Therefore, it is hard to deduce whether mammalian Bcnt is involved in intracellular trafficking. Recently, we have learned that a knockdown of Drosophila Bcnt by RNA interference resulted in lethality at the larval stage (S. Iwashita, S. Gotoh, K. Otsu, R. Kuwahara, and R. Ueda, personal communication). These evidences and studies on mouse Bcnt by antibody-blocking experiments (Diekwisch and Luan 2002) suggest that Bcnt has an indispensable role, but the function of either Bcnt or p97Bcnt is still not obvious.

Gene bcnt^p97 was created by gene duplication followed by BovB/RTE insertion and exonization in one of the duplicates. This type of evolutionary process was already described, for example, in the Jingwei gene of Drosophila, where gene duplication of alcohol dehydrogenase followed by an exon insertion created a new dehydrogenase gene with altered substrate specificity (Zhang et al. 2004). So far, four models have been proposed to account for the processes leading to creation of a novel gene with a new function, namely the neofunctionalization model (Ohno 1970), subfunctionalization model (Force et al. 1999), adaptation model (Hughes 2002), and the recently proposed adaptive radiation model (Francino 2005). Although the adaptive radiation model was proposed based on postulation of large and selected amplification of the best preadapted genes for creation of new gene function, the process of bcnt^p97 creation may fit better its concept rather than those of the other models. TEs can speed up the natural mutagenesis process tremendously in the evolution (Makalowski 2003). And our results on the time frame of bcnt^p97 creation strongly suggest that this early process might have coincided with the horizontal transfer of an external RTE-1 element and/or rapid expansion of the elements. Therefore, the insertion of RTE-1 into the gene duplicates in the era had a great potential for creation of novel gene with a new function. First, while bcnt is a well-conserved gene, insertion of a BovB/RTE in this locus and creation of bcnt^p97 occurred only in the lineage of ruminants. It is unlikely that the function of bcnt^p97 coexisted with bcnt prior to bcnt^p97 creation. Therefore, the evolution of new function by recruitment of AP-END of a BovB/RTE as a major exon is more likely. While the two models of subfunctionalization and adaptation require the multiple functions to coexist prior to gene duplication, the adaptive radiation model allows the evolution of new functions from existing ones. Secondly, insertion of a TE, for example, a LINE into a gene generally deteriorates biological activities and significantly suppresses its transcription (Han, Szak, and Boeke 2004). But bcnt^p97 exonized an AP-END of BovB/RTE as a novel gene. These processes of bcnt^p97 creation might be coincident with the concept of the adaptive radiation model that allows a period of natural selection for the duplication before new function evolves, while the other models consider gene duplication to be neutral. The adaptive radiation model is also consistent with the finding that a substantial acceleration in the evolution was observed in the paralogs after duplication (data not shown) as was observed for other cases (Jordan, Wolf, and Koonin 2004; Lynch and Katju 2004). In addition, the ruminant bcnt cluster separately from other animals in the phylogenetic tree (fig. 5). For a more complete picture of the recruitment process and the BovB/RTE exonization, we are further pursuing a retrospective analysis of the changes that led to the creation of this gene.

In conclusion, creation of bcnt^p97 had most likely occurred in an ancestral ruminant before the radiation of its species. The comparative analysis reveals major changes in the genomic regions that show characteristic footprints of TEs. Therefore, this study provides insight into the local type of genomic rearrangements by which paralogous genes are formed, shaped, and acquire new function on the background of the global genomic evolution associated with the expansion of TEs.

	Note Added in Proof

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 
After acceptance of this paper, we have noticed a gene similiar to bcnt^p97 (GenBank accession number LOC514131) that is located 7.6 kb upstream of bcnt^p97 on bovine chromosome 18 and potentially encodes a protein with 664 amino acids.

	Supplementary Material

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 
Supplementary Materials and Methods, tables S1 and S2, and figures S1–S3 are available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/).

	Acknowledgements

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 
We thank the Sabah Wildlife Department of Malaysia for the permit to use wild animals for this study, H. Yasue for isolation of pig BAC clones, and Y. Kurosawa and Y. Takagaki for camel spleen and pig DNA, respectively. We gratefully acknowledge M. P. Francino for comments and discussion of the process of bcnt^p97 creation in adaptive radiation model. We are also grateful to three anonymous reviewers, T. Yamada, M. Lynch, M. Bazter, and I. D. Hincken for the useful discussion. We thank M. Tanio for three-dimensional modeling of bcnt^p97 RTE exon, H. B. Tamate, K. Tsuge, S. Kaneko, and N. Koike for the technical advice, and the former president Y. Nagai for the encouragement. The group on "Physiological and ecological investigation of mouse deer and their conservation in Malaysia and Thailand" led by K.F. was supported by the Ministry of Education, Culture, Sports, Science and Technology, Japan, and the present work was partially supported by its grant-in-aid (#13575027).

	Footnotes

 
David Irwin, Associate Editor

	References

TOP Abstract Introduction Materials and Methods Results Discussion Note Added in Proof Supplementary Material Acknowledgements References

 

Beernink, P. T., B. W. Segelke, M. Z. Hadi, J. P. Erzberger, D. M. Wilson III, and B. Rupp. 2001. Two divalent metal ions in the active site of a new crystal form of human apurinic/apyrimidinic endonuclease, Ape1: implications for the catalytic mechanism. J. Mol. Biol. 307:1023–1034.[CrossRef][ISI][Medline]

Breukelman, H. J., P. A. Jekel, J.-Y. Dubois, P. P. M. F. A. Mulder, H. W. Warmels, and J. J. Beintema. 2001. Secretory ribonucleases in the primitive ruminant chevrotain (Tragulus javanicus). Eur. J. Biochem. 268:3890–3897.[ISI][Medline]

Brosius, J. 2005. Echoes from the past—are we still in an RNP world? Cytogenet. Genome Res. 110:8–24.[CrossRef][ISI][Medline]

Diekwisch, T. G., and X. Luan. 2002. CP27 function is necessary for cell survival and differentiation during tooth morphogenesis in organ culture. Gene 287:141–147.[CrossRef][ISI][Medline]

Endo, H., J. Kimura, M. Sasaki, M. Matsuzaki, H. Matsubayashi, K. Tanaka, and K. Fukuta. 2002. Functional morphology of the mastication muscles in the lesser and greater mouse deer. J. Vet. Med. Sci. 64:901–905.[CrossRef][ISI][Medline]

Force, A., M. Lynch, F. B. Pickett, A. Amores, Y.-L. Yan, and J. Postlethwait. 1999. Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151:1531–1545.[Abstract/Free Full Text]

Francino, M. P. 2005. An adaptive radiation model for the origin of new gene functions. Nat. Genet. 37:573–577.[CrossRef][ISI][Medline]

Fritz, G., and B. Kaina. 1999. Phosphorylation of the DNA repair protein APE/REF-1 by CKII affects redox regulation of AP-1. Oncogene 18:1033–1040.[CrossRef][ISI][Medline]

Han, J. S., S. T. Szak, and J. D. Boeke. 2004. Transcriptional disruption by the L1 retrotransposon and implications for mammalian transcriptomes. Nature 429:268–274.[CrossRef][Medline]

Hassanin, A., and E. J. Douzery. 2003. Molecular and morphological phylogenies of ruminantia and the alternative position of the moschidae. Syst. Biol. 52:206–228.[CrossRef][ISI][Medline]

Hon-Nami, K., S. Ueno, H. Endo, H. Nishimura, H. Igarashi, L. David, and S. Iwashita. 2004. A novel Giraffidae-specific interspersed repeat with a microsatellite, originally found in an intron of a ruminant paralogous p97bcnt gene. Gene 340:283–290.[CrossRef][ISI][Medline]

Hughes, A. L. 2002. Adaptive evolution after gene duplication. Trends Genet. 18:433–434.[CrossRef][ISI][Medline]

Iwashita, S., T. Nobukuni, S. Tanaka, M. Kobayashi, T. Iwanaga, H. B. Tamate, T. Masui, I. Takahashi, and K. Hashimoto. 1999. Partial nuclear localization of a bovine phosphoprotein, BCNT, that includes a region derived from a LINE repetitive sequence in Ruminantia. Biochim. Biophys. Acta 1427:408–416.[Medline]

Iwashita, S., N. Osada, T. Itoh et al. (11 co-authors). 2003. A transposable element-mediated gene divergence that directly produces a novel type bovine Bcnt protein including the endonuclease domain of RTE-1. Mol. Biol. Evol. 20:1556–1563.[Abstract/Free Full Text]

Jordan, I. K., Y. I. Wolf, and E. V. Koonin. 2004. Duplicated genes evolve slower than singletons despite the initial rate increase. BMC Evol. Biol. 4:22.[CrossRef][Medline]

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.[CrossRef][ISI][Medline]

Kordis, D., and F. Gubensek. 1998. Unusual horizontal transfer of a long interspersed nuclear element between distant vertebrate classes. Proc. Natl. Acad. Sci. USA 95:10704–10709.[Abstract/Free Full Text]

Lev-Maor, G., R. Sorek, N. Shomron, and G. Ast. 2003. The birth of an alternatively spliced exon: 3' splice-site selection in Alu exons. Science 300:1288–1291.[Abstract/Free Full Text]

Liu, W.-M., W.-M. Chu, P. V. Choudary, and C. W. Schmid. 1995. Cell stress and translational inhibitors transiently increase the abundance of mammalian SINE transcripts. Nucleic Acids Res. 23:1758–1765.[Abstract/Free Full Text]

Long, M. 2001. Evolution of novel genes. Curr. Opin. Genet. Dev. 11:673–680.[CrossRef][ISI][Medline]

Lynch, M., and V. Katju. 2004. The altered evolutionary trajectories of gene duplicates. Trends Genet. 20:544–549.[CrossRef][ISI][Medline]

Makalowski, W. 2000. Genomic scrap yard: how genomes utilize all that junk. Gene 259:61–67.[CrossRef][ISI][Medline]

———. 2003. Not junk after all. Science 300:1246–1247.[Abstract/Free Full Text]

Malik, H. S., and T. H. Eickbush. 1998. The RTE class of non-LTR retrotransposons is widely distributed in animals and is the origin of many SINEs. Mol. Biol. Evol. 15:1123–1134.[Abstract]

Modi, W. S., D. S. Gallagher, and J. E. Womack. 1996. Evolutionary histories of highly repeated DNA families among the Artiodactyla (Mammalia). J. Mol. Evol. 42:337–349.[ISI][Medline]

Morales, J. F., E. T. Snow, J. P. Murnane. 2003. Environmental factors affecting transcription of the human L1 retrotransposon. II. Stressors. Mutagenesis 18:151–158.[Abstract/Free Full Text]

Nekrutenko, A., and W.-H. Li. 2001. Transposable elements are found in a large number of human protein-coding genes. Trends Genet. 17:619–621.[CrossRef][ISI][Medline]

Nikaido, M., A. P. Rooney, and N. Okada. 1999. Phylogenetic relationships among cetartiodactyls based on insertions of short and long interpersed elements: hippopotamuses are the closest extant relatives of whales. Proc. Natl. Acad. Sci. USA 96:10261–10266.[Abstract/Free Full Text]

Nobukuni, T., M. Kobayashi, A. Omori et al. (12 co-authors). 1997. An Alu-linked repetitive sequence corresponding to 280 amino acids is expressed in a novel bovine protein, but not in its human homologue. J. Biol. Chem. 272:2801–2807.[Abstract/Free Full Text]

Ohno, S. 1970. Evolution by gene duplication. Allen and Unwin, Springer-Verlag, London.

Poulter, R., M. Butler, and J. Ormandy. 1999. A LINE element from the pufferfish (fugu) Fugu rubripes which shows similarity to the CR1 family of non-LTR retrotransposons. Gene 227:169–179.[CrossRef][ISI][Medline]

Suzuki, K., S. Asakawa, M. Iida, S. Shimanuki, N. Fujishima, H. Hiraiwa, Y. Murakami, N. Shimizu, and H. Yasue. 2000. Construction and evaluation of a porcine bacterial artificial chromosome library. Anim. Genet. 31:8–12.[CrossRef][ISI][Medline]

Szemraj, J., G. Plucienniczak, J. Jaworski, and A. Plucienniczak. 1995. Bovine Alu-like sequences mediate transposition of a new site-specific retroelement. Gene 152:261–264.[CrossRef][ISI][Medline]

Takahashi, I., T. Nobukuni, H. Ohmori, M. Kobayashi, S. Tanaka, K. Ohshima, N. Okada, T. Masui, K. Hashimoto, and S. Iwashita. 1998. Existence of a bovine LINE repetitive insert that appears in the cDNA of bovine protein BCNT in ruminant, but not in human, genomes. Gene 211:387–394.[CrossRef][ISI][Medline]

Tatusova, T. A., and T. L. Madden. 1999. Blast 2 sequences, a new tool for comparing protein and nucleotide sequences. FEMS Microbiol. Lett. 174:247–250.[CrossRef][ISI][Medline]

Tell, G., G. Damante, D. Caldwell, and M. R. Kelley. 2005. The intracellular localization of APE1/Ref-1: more than a passive phenomenon? Antioxid. Redox Signal. 7:367–384.[CrossRef][ISI][Medline]

Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876–4882.[Abstract/Free Full Text]

Wen, Y., and D. M. Irwin. 1999. Mosaic evolution of ruminant stomach lysozyme genes. Mol. Phylogenet. Evol. 13:474–482.[CrossRef][ISI][Medline]

Wisniewski, T. P., C. L. Tanzi, and J. G. Gindhart. 2003. The Drosophila kinesin-I associated protein YETI binds both kinesin subunits. Biol. Cell. 95:595–602.[CrossRef][ISI][Medline]

Yang, Z. 1997. PAML: a program package for phylogenetic analysis by maximum likelihood. Comput. Appl. Biosci. 13:555–556.[Free Full Text]

Zhang, J., A. M. Dean, F. Brunet, and M. Long. 2004. Evolving protein functional diversity in new genes of Drosophila. Proc. Natl. Acad. Sci. USA 101:16246–16250.[Abstract/Free Full Text]

Zupunski, V., F. Gubensek, and D. Kordis. 2001. Evolutionary dynamics and evolutionary history in the RTE clade of non-LTR retrotransposons. Mol. Biol. Evol. 18:1849–1863.[Abstract/Free Full Text]

Accepted for publication December 22, 2005.

CiteULike    Connotea    Del.icio.us    What's this?

This Article Abstract FREE Full Text (PDF) Supplementary Material All Versions of this Article: 23/4/798    most recent msj088v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My P