Identification of an Xiap-Like Pseudogene on Mouse Chromosome 7

The most thoroughly characterized mammalian IAP is XIAP/BIRC4, which can inhibit caspases 9, 3 and 7, but may also regulate apoptosis through interactions with other proteins such as Smac/DIABLO, HtrA2/Omi, XAF1, TAK1, cIAP1, and cIAP2. High throughput sequencing of the mouse genome revealed the existence of a gene resembling Xiap/Birc4 on mouse chromosome 7. To confirm the existence of this gene, and to determine its functional significance, we performed Southern and Northern blot analysis. This showed the presence of the Xiap-like gene in both wild-type and Xiap gene knock-out mice, but the corresponding mRNA was not detected in any tissues examined by Northern blot. Analysis of the gene sequence in all three possible reading frames predicts that expression of this gene would not give rise to a full-length protein, but only non-functional truncated polypeptides. Because its nucleotide sequence is 92% identical to Xiap, but it has no introns corresponding to those of Xiap, we conclude that Xiap-ps1 is a pseudogene generated by retro-transposition of a spliced Xiap message to chromosome 7.

Surprisingly, although XIAP is the most potent caspase inhibitor of the IAPs [6], the phenotype of Xiap knockout mice is very mild [7] and humans harbouring XIAP/BIRC4 mutations have immune system defects due to abnormal NK cell function, but are otherwise normal [8].
We were alerted (Anthony Uren pers. comm.) to the presence of a Xiap-like sequence found on mouse chromosome 7 during sequencing of the mouse genome (see: http://apr2006.archive.ensembl.org/ Mus_musculus/domainview?domainentry = IPR001370). To determine whether this gene was functional, and whether redundancy with Xiap might explain the subtle phenotype of Xiap KO mice, we analysed the sequence of this putative Xiap-like gene, confirmed its existence by Southern blot, and used Northern analysis to determine whether it was expressed.

Ethics Statement
All mouse work was done according to the requirements of La Trobe University Animal Ethics Committees with ethics approval number: AEC 09-01-B. Animals were sacrificed using CO2 asphyxiation and the appropriate organs harvested.

Tissue Samples
Tissue samples (liver, brain, lung, spleen, heart and intestine) were harvested from wild-type and Xiap 2/2 C57BL/6 mice and were snap frozen in liquid nitrogen and stored at 280uC until required.

Southern Blot Analyses
Digested DNA samples were separated by electrophoresis on 1% agarose TAE gels containing 0.2 mg/ml ethidium bromide. Afterwards, the gel was denatured in 1.5 M NaCl/0.5 M NaOH for 45 min and then neutralised in 1 M Tris (pH 7.4)/1.5 M NaCl. Separated DNA was transferred onto Zeta-probe (BioRad) nitrocellulose membrane by overnight capillary transfer and fixed by baking at 80uC for 2 hr. The membrane was pre-hybridised for 30 min at 65uC in Rapid-Hyb buffer (Amersham) followed by hybridisation with 32 P labelled mXiap DNA probe at 65uC for 2 hr in Rapid-Hyb buffer according to the manufacturer's specifications. The membrane was washed twice at 65uC for 30 min in 0.3X SSC. The hybridised probe was detected using a Typhoon phosphoimager (Amersham).
The DNA probe was obtained by digesting Xiap cDNA to make a 510 bp fragment identical to bases 524 to 1034 of the cDNA of Xiap, a region within exon 2 and encoding amino acids 105 to 275 of the protein. Alignment of the probe to Xiap-ps1 showed 92% identity, indicating that the probe would be able to hybridise to both genes at high stringency.

RNA Extraction and Northern Blot Analyses
To isolate RNA, frozen tissue samples were homogenised in Trizol (Invitrogen) and RNA purified according to the manufacturer's instructions. Total RNA (10-15 mg) was separated on 1.2% agarose formaldehyde gels, transferred to Hybond-N (Amersham) membrane by capillary transfer and fixed by baking at 80uC. Membranes were pre-hybridised for 30 min at 65uC in Rapid-Hyb buffer followed by hybridisation with the 32 P labelled mXiap DNA probe at 65uC for 2 hr. The final wash was three times for 20 min at 65uC in 0.3X SSC+0.5% SDS, and the hybridised probe was detected using a Typhoon phosphorimager (Amersham).

Results
Identification of a Xiap-Like Gene (Xiap-ps1) on Chromosome 7 On mouse chromosome 7 band B3 location 37,599,271-37,600,785 is a sequence with 92% nucleotide identity to Xiap cDNA (Fig. 1a,b). To confirm the existence of the Xiap-ps1 gene in the C57BL/6 mouse genome we performed Southern blot analysis of genomic DNA from both wild-type (WT) and Xiap deleted mouse embryonic fibroblasts (MEFs), as well as from tissues from C57BL/6 WT and Xiap deleted mice.
As indicated in Fig. 1a, digestion of WT DNA with BamHI, EcoRI, HindIII and XbaI restriction enzymes is predicted to produce fragments of the mouse Xiap gene of 2.1 kb, 6.8 kb, 6 kb and 2 kb, respectively, that would be able to hybridise to an Xiap cDNA probe containing the coding region of exon II. In the Xiap mutant mice, exon II of Xiap has been deleted by homologous recombination, and therefore genomic DNA isolated from these mice does not contain any Xiap sequences that are capable of hybridising with the probe.
Any bands that hybridise at high stringency (65uC; 0.3x SSC) to the Xiap probe in the DNA from Xiap knockout mice indicate the presence of another gene similar to Xiap. As seen in Fig. 2, hybridisation of the probe to WT samples not only gave rise to fragments of the expected size for Xiap, but also additional bands that were also seen in the DNA from the Xiap-deleted mice. This confirms the presence of a Xiap-like gene that we have designated Xiap-ps1. These bands of 4.2 kb, 3.4 kb, 6.9 kb and 4.4 kb in samples digested with BamHI, EcoRI, HindIII and XbaI, respectively, are consistent with those predicted from the digestion of the Xiap-ps1 gene sequence on chromosome 7 in Genbank (http://apr2006.archive.ensembl. org/Mus_musculus/domainview?domainentry = IPR001370) with the same restriction enzymes (Fig. 1a).

Xiap-ps1 Is Not Expressed
To determine whether Xiap-ps1 is expressed we performed RNA Northern blot analyses. Tissues from WT and Xiap deleted mice were harvested and total RNA was isolated. The RNA samples were separated on denaturing formaldehyde agarose gels, transferred to membrane and probed with the same 32 P-labelled probe that was   Figure 1. The only band detected (,6.6 kb) is the reported size for Xiap mRNA, and was only found in wild-type tissues and not in the Xiap knockout tissues. A b-actin probe was used to re-probe the same blots to show relative loading of lanes. doi:10.1371/journal.pone.0008078.g003 used in the genomic Southern analyses. As shown in Fig. 3, in all WT tissue samples a single band of approximately 6.6 kb was detected, consistent with previous reports of the size of mouse Xiap mRNA [7,9]. In contrast, in samples from Xiap knockout mice no bands were detected, indicating that the Xiap-ps1 mRNA is not expressed at detectable levels in these tissues in vivo.

Xiap-ps1 Bears Premature Stop Codons
Analysis of the Xiap-ps1 nucleotide sequence showed that it does not code for a full-length protein due to the presence of premature stop codons. Translation of the gene sequence in all three possible reading frames gives rise to several truncated polypeptide sequences (Fig. 4). This indicates that even if Xiap-ps1 were transcribed, the presence of these premature stop codons would prevent translation of a functional IAP protein.

Discussion
From the Southern analyses (Fig. 2) we have confirmed that a novel Xiap-like gene exists on mouse chromosome 7. As we detected the presence of the Xiap-ps1 gene in genomic DNA samples from Xiap2/2 tissues, the bands detected were not the result of the probe hybridising to the Xiap gene on the X chromosome. We failed to detect any evidence by Northern analysis that Xiap-ps1 is expressed in tissues in which Xiap is clearly expressed at the mRNA level. Analysis of the nucleotide sequence showed that Xiap-ps1 gene is devoid of intronic sequences found in Xiap and analysis of the various splicing variants showed that Xiap-ps1 shows similarity to the regions in common to all variants, starting 7 nucleotides upstream of the initiation codon in exon 2 of Xiap (Fig. 1b). This suggests that Xiap-ps1 has arisen from retrotransposition of a processed Xiap mRNA to chromosome 7. We hypothesise that a retroviral infection occurred in mouse germ cells that allowed insertion of reverse-transcribed spliced Xiap mRNA into chromosome 7. A similar event has been shown to have occurred during evolution of the great apes, leading in that case to a transcribed and translated product, BIRC8 (ILP-2), from a single-exon gene [10].
Examination of the flanking genomic sequences of Xiap-ps1 using the Transfac database revealed several transcriptional elements, the closest to the gene being a retroviral TATA box and a CAP signal for transcription initiation 1692 and 1629 bp upstream of the initiating methionine respectively. The presence of viral elements supports our hypothesis that retrotransposition of the Xiap mRNA occurred.
Translation of the sequence for Xiap-ps1 indicates that no full length protein could be produced. Although three of the larger peptide sequences were very similar to regions of XIAP, it is likely that even if these peptides were expressed within cells they would not be able to function as inhibitors of apoptosis.
Author Contributions Figure 4. Alignment of translated pseudogene to WT XIAP protein. Three small methionine-initiated peptides are predicted from translation of the pseudogene sequence in all three frames. Functional domains are highlighted by background shading. XIAP-ps1 peptides encompassing BIR 1 and the RING domain are translations from the first frame, whereas the peptide containing BIR3 is a translation in the 3 rd frame. Alignment of these peptides to WT XIAP shows significant similarity with regions of XIAP, as shown in pale grey lettering, but no full length protein could be produced. doi:10.1371/journal.pone.0008078.g004