The yak (Bos grunniens) is one of the most remarkable domestic animals and an iconic symbol of Qinghai-Tibetan plateau, which provides the basic resources (meat, milk, transportation, dung for fuel and hides for tented accommodation) necessary for the Tibetans and other nomadic pastoralists in high altitude environments ranging from 3,000 to 5,500 m above the sea level [1,2], and has been reported as a typical four seasons grazing ruminant in the area [3]. In order to adapt to the harsh environments and atmosphere with severe cold, less oxygen, strong ultra-violet (UV) radiation, and poor forage resources, The yak has evolved special adaptations in physiology, nutrient metabolism and foraging [4]. Due to long exposure to these environments, the genome is continuously affected by exogenous and endogenous genotoxic agents, as ionizing radiation, drugs, and byproducts of normal cellular metabolism that generate reactive oxygen species (ROS) leading to mainly non bulky DNA lesions [5], which may lead to repeated, lost, mutation of DNA and add more alleles because of complex or rearrangement. The yak has strong adaption in such harsh conditions, the apurinic/apyrimidinic repair enzyme 1 (APEX1) may play an important role in this process. The apurinic/apyrimidinic (AP) site is incised by an essential enzyme known as apurinic/apyrimidinic endonuclease 1 (APE1) or APEX1 in animal [6]. The damaged DNA is repaired by many enzymes, among the Base Excision Repair (BER) enzymes, AP endonucleases play a crucial role in BER because they recognize the abasic residues and hydrolyze the phosphodiesterbond 5′-deoxyribosephosphate termini to the AP sites, leaving a gapped DNA intermediate with an extendable 3′-hydroxyl end [7]. APEX1 functional activation is a consequence of different stimuli that may generate both physiological and toxic oxidative stress conditions or increase the intracellular cAMP levels leading to different outcomes [8,9]. Therefore, recognition of AP sites is fundamental to cell/organism survival. Protein upregulation is always associated with an increase in both redox and AP endonuclease activity, followed by an increase in cell resistance toward oxidative stress and DNA damaging agents [10], strengthening the conclusion that an upregulation of APEX1 protein levels has profound biological consequences. However, this work was the first time to report the nucleotide sequence, amino acid sequence and bioinformatics of APEX1 in domestic yak and the findings may have important implications for understanding adaptation to high altitude in other species and for hypoxia-related diseases in humans.

Samples and materials

Fresh tissues of liver, kidney, spleen, lung and testis were obtained immediately after slaughtering from five randomly selected adult male yaks in the slaughterhouse of Tianzhu county (Wuwei city, Gansu province, China), and the tissues were instantly stored at -80°C.

RNA Isolation and cDNA synthesis

Total RNA was extracted from yak liver tissues by using RNAiso Plus (Takara Code No. 9109, Dalian, China) following manufacturer’s instruction and was further used for cDNA synthesis. RNA was quantified by NanoDrop-8000 (Thermo, Waltham, MA, USA) and RNA integrity was assessed on denaturing formaldehyde agarose gel (1%) electrophoresis (Biowest Regular Agarose, No.111860, Castropol, Spain). 2 μg total RNA was applied to reverse transcription to single-stranded cDNA using BioTeke Thermo RT Kit (Bioteke, Cat No.PR6601, Beijing, China), the reverse transcription PCR reaction was 48°C for 50 min, followed by 70°C for 10 min. The total 20 μL cDNA synthesis reaction contains 2 μL total RNA, 1 μL Oligo dT or Random Primer (50 μM), dNTP Mixture (10 μM), Thermo M-MLV (200 U/μL), RNase Inhibitor (40 U/μL), 4μL 5× first-strand buffer and 10 μL of ultrapure millipore water (Invitrogen, Carlsbad, CA, USA).

Synthesis and confirmation of partial cDNA of APEX1 gene

The conserved sequences were found based on homology from Bos taurus, Camelus dromedaries and Sus scrofa by using Nucleotide BLAST and Primer-BLAST available from National Center for Biotechnology Information (NCBI). The primers of yak APEX1 were designed in the conserved sequences by Premier 5 and Oligo 7.0 softwares [11]. The special primers used for initial screening and obtaining partial yak APEX1 cDNA were Forward 5’-CTCACCCAGTGGCAAATCAG-3’ and Reverse 5’-TGTCACACARTGCAGGCAA-3’. The PCR reaction mixture (30 μl) contains 2 μL of template cDNA, 1.5 μL (10 μM) of each primers set, 15 μL 2× Power Taq PCR MasterMix, (BioTeke Cat No.71182, Beijing, China) and 10 μL of ultrapure millipore water and was carried out by Bio-Rad PCR (BIO-RAD, Hercules, CA, USA). The PCR amplification conditions were as follows: initial denaturation at 95°C for 5 min followed by 35 cycles at 95°C for 1 min, 52°C for 1 min and 72°C for 1 min. Final extension step was carried out at 72°C for 10 min. The products of PCR reaction were electrophoresed on agarose gel (1.5%) using DNA Ladder DL2000 (Genstar, Cat No.TZM122-01, Chendu, China). The 735 bp fragment from PCR reaction was eluted and purified using BioTeke Gel Extraction kit (BioTeke Lot No.G51101, Beijing, China). Purified fragments were quantified and validated on agarose electrophoresis before sequencing. PCR fragment was sequenced by BGitech Company, Beijing, China.

Rapid amplification of cDNA ends (RACE), cloning and sequencing

Rapid-amplification of cDNA ends (RACE-PCR) was used to identify and isolate the 5’- and 3’-ends of APEX1 using RACE kits (Invitrogen No.18373019/18374058, Carlsbad, CA, USA). Yak total RNA was annealed with 5’- and 3’-ends adaptor primers, and reversely transcribed respectively to respective 5’- and 3’-cDNAs. The resulting single stranded 5’- and 3’-cDNAs were used as templates in PCR. For 5’ RACE, the 5’-end adaptor primer (Forward primer 5’-GGAAATAAAAAGAATGCTGGCT-3’) and APEX1 5R (Reverse primer 5’-CCAGCACCTCTACCCCAACAC-3’) were used. For 3’-RACE, APEX1 3F primer (Forward primer 5’-CCAATGCCATAGGAGACTTTGAGCGG-3’) and the 3’-end adaptor primer (Reverse primer 5’-GCATCATCTTCAGCCACAATCACTCGG-3’) were used. The cycling program was set for five cycles of 94°C, 5 min; 5 cycles of 94°C 15 s, 70°C 15 s, 72°C 3 min; 5 cycles of 94°C 15 s, 68°C 15 s, 72°C 3 min; 5 cycles of 94°C 15 s, 65°C 15 s, 72°C 3 min; 25 cycles of 94°C 15 s, 60°C 15 s, 72°C 3 min; 1 cycle of 72°C, 5 min. The RACE PCR products were purified and cloned into pMD-18 vector (Takara Code No.6011, Dalian, China) to confirm the sequences of the 5’- and 3’-ends, respectively. The full-length cDNA PCR product was cut with the restriction enzymes, purified, and ligated with T4 DNA ligase into a pMD-18 vector. Competent E. coli DH5α cells were transformed with the plasmids and selected by means of antibiotic resistance. The sequence of the inserted full-length cDNA was confirmed with DNA sequencing.

Analysis and alignment of cDNA sequence

The sequence of yak APEX1 was analyzed for nucleotide, protein translation, sequence alignment and comparison with other mammalian species [12]. The sequenced DNA fragment was translated by EditSeq of DNASTAR program [13] and the deduced yak APEX1 amino acid sequence was compared in the NCBI protein database using the BLASTP algorithm [14]. DNA and protein homologies were analyzed by using NCBI BLAST program via NCBI web servers. SignalP 4.1 server was used to predict the presence and location of signal peptide cleavage sites in amino acid sequence of yak APEX1 (http://www.cbs. dtu.dk/services/SignalP/) [15]. TMHMM Server v.2.0 server was used to predict the transmembrane helices in protein (http://www.cbs.dtu. dk/services/TMHMM/) [16]. ProtScale software was used to analyze hydrophilic/hydrophobic of yak APEX1 (http://web.expasy.org/cgibin/ protscale/protscale.pl) [17]. An amino acid scale was defined by a numerical value assigned to each type of amino acid. The most frequently used scales were the hydrophobicity or hydrophilicity scales and the secondary structure conformational parameters scales, but many other scales based on different chemical and physical properties of the amino acids also exist [18].

Secondary structure prediction and multiple sequence alignment of yak APEX1

CLCbio free Genome workbench (CLC Genome workbench, version 7) was used to construct the secondary structure of yak APEX1 mRNA sequence with the free energy minimization algorithm. RNA tertiary structures was characterized by secondary structural elements based on hydrogen bonds within the molecule that form several recognizable “domains” of secondary structure like stems, hairpin loops, bulges and internal loops [19,20]. A poly (A) tail and poly (A) signal for the yak APEX1 mRNA was predicted based on free minimization algorithm of CLCbio. The predicted molecular weight, isoelectric point (pI) and charges of the protein APEX1 were estimated. The amino acid sequence of yak APEX1 was used as template to identify homologous mammalian sequences in PSI-BLAST.

Yak APEX1 protein 3D structure prediction

The 3D structure of yak APEX1 was predicted by Swiss-model server base on homology structure modeling [21,22]. The 3D structure of human APEX1 (PDB: 1e9n.1.A) at X-ray, 2.20 Å resolution with 90.25% sequence identity was used as a template to predict the 3D structure of yak APEX1. The similarities between modeled APEX1 structure and human APEX1 were superimposed by using Pymol software (delino Scientific) [23]. The quality of the superimposed 3D structures was assessed using PDBe on EMBL-EBI server.

Real-time PCR assays for APEX1 gene expression in yak tissues

The expression levels of APEX1 were determined by quantitative Real-time PCR. 200 ng total RNA was used for the synthesis of singlestranded DNA (cDNA) with BioTeke Thermo RT Kit (Bioteke, Cat No.PR6601, Beijing, China). PCR was performed using 2 μl of cDNA in a 25 μl reaction volume with ABI 7300 (ABI system, Foster, CA, USA) Real-time System. SYBR Premix Ex TaqTMII and specific primers were used in each reaction. The forward and reverse primers were designed with the Primer 5.0 based on the APEX1 cDNA sequences. Primer sequences used were as follows, Forward: 5’-AAGAGTAAGGCAGGAGCGAAAA-3’, Reverse: 5’-ATCCACATTCCACGAGCAGA-3’, product size 138 bp), GAPDH (Forward: 5’-AAGTGGGGTGATGCTGGTG-3’, Reverse: 5’-GCTGACAATCTTGAGGGTGTTG-3’, product size 189 bp, Genbank No. NM_008084). The results were expressed relative to liver as calibrator and using the GAPDH as housekeeping gene. ABI7300 Real-time PCR system was used, and the 2 step standard procedure for amplifying the samples was applied. Stage 1: for denaturation, Reps 1, 95°C for 30 s. Stage 2: PCR reaction, Reps 40, 95°C for 5 s and 55°C for 34 s. The final step was the dissociation stage. All Real time PCR reactions were performed in triplicate. The results were calculated applying the 2^-ΔΔCT method [24].

APEX1 cDNA sequence

Based on sequence homology, the designed primers amplified a single 735 bp fragment of APEX1 gene. The RACE technology was used and a full-length cDNA fragment of 1391 bp was successfully obtained. The sequence analysis showed 957 bp open reading frame (ORF) of APEX1 gene and consisted of 125 bp and 309 bp corresponding to the 5′- and 3′-terminal untranslated regions (UTR), respectively (Figure 1). The sequence was submitted to Genbank and got the accession number KF611690. The initiation codon (ATG) and end codon (TGA) were shown at position 126 bp and 1084 bp, respectively. The poly (A) tail signal peptide and poly (A) tail were shown at the positions of 1346 and 1365 bp respectively.

Figure 1: Complete nucleotide sequence encoding Bos grunniens APEX1 gene submitted to NCBI GenBank (accession number: KF611690).

Analysis and functional anotation of the secondary structure

The molecular analysis of the 317 amino acid sequence of APEX1 showed that this protein has a molecular weight of 35.1KDa, pI 7.82, and the molar extinction coefficient 56270% ± 5%. The predicted protein contain 110 charged amino acids (34.6%), 105 hydrophobic (33.0%), 40 acidic (12.6%), 42 basic (13.2%) and 77 polar amino acids (24.2%). The comparison between the predicted amino acid sequence of APEX1 and the sequences from different organisms were carried out. A prediction of the secondary structure analysis of APEX1 suggested that 71 amino acids was composed of 8 α helix and 16 β sheets, 183 amino acids was formed of random coil in APEX1 protein (Figure 2), the potential phosphorylation sites were analyzed based on the CBS Prediction Servers (NetPhos 2.0 Server), and S26, S54, S56, S123 and S298, T19 and T233, Y45, Y128, Y257 and Y262 were predicted as potential phosphorylation sites. S54, S56, S123 and Y128 have a score of above 0.95 indicating a very likely phosphorylation site. The CLC Genomics Workbench 7 program was used for annotating the functional sites of APEX1 sequence. The DNA binding sites, putative catalytic sites, putative phosphate binding sites, ion binding sites and active sites were shown in Figure 2.

Figure 2: The predicted secondary structure and annotated function sites of the APEX1 sequence using CLC Genomics Workbench 7 program.

Alignment of APEX1s amino acid sequence

The comparison between the predicted amino acid sequence of yak APEX1 with the most similar APEX1 sequences from Bos grunniens (KF611690), Bos taurus (NM_176609.3), Sus scrofa (NM_001139471.1), Camelus dromedaries (HM209828.2), Pan troglodytes (DQ977332.1), Homo sapiens (AAP36916.1), Mus musculue (NM_009687.2) and Canis lupus families (NP_001138591.1) indicated that the amino acid sequences of APEX1 were aligned with 8 different mammalians APEX1 by ClustalW program. The highest similarity was found with B. taurus (96%), S. scrofa (93%), C. dromedaries (92%), P. troglodytes (91%), H. sapiens (90%), M. musculus (90%) and C. lupus (90%), respectively (Figure 3). Such high similarity proposed a close evolutionary relationship in these mammalian species.

Figure 3: Multiple sequence alignment of amino acid sequences of the yak APEX1s with seven mammalian species. (Mus musculue, Homo sapiens, Pan troglodytes, Bos taurus, Sus scrofa, Canis lupus familiaris and Camelus dromedaries). #conserved cysteine; *catalytic and enzymatically important residues; residues involved in DNA backbone interactions.

Prediction and alignment 3D structure of yak APEX1

The APEX1 protein 3D structure was predicted by Swiss-Pdb Viewer 4.1.0 (Figure 3A). Ribbon structure was represented of the yak APEX1. The α/β-sandwich was formed by the packing of two sixstranded β sheets surrounded by α helices, with strand order β5-β6-β4- β3-β16-β15 and β7-β8-β9-β10-β14 from domains 1 and 2, respectively. The α-helices and β-strands were numbered sequentially from the N-terminus (Figure 4A and 4B). The active sites were lies in a pocket at the top of the α/β-sandwich and were surrounded by loop regions. The structural similarity between yak and human APEX1 were studied by superimposing their structures using the Pymol program (http:// pymol.sourceforge.net). The folds and topology of modeled yak APEX1 was very similar to human APEX1 (Figures 4B). Within the active site, the imidazole ring of His 309 interacts with the carboxylate of Asp 283, which in turn forms a hydrogen bond with Thr 265 (Figure 4C). The side chains Tyr 171 and Glu 96 were hydrogen bonded, as were Asn 68, Asp 210 and the main chain amides of Asp 70 and Asn 212, which together form a hydrogen bonding network. The first metal binding site (metal A) was coordinated by the carboxylates of residues Asp 70 and Glu 96. The second metal binding site (site B) was composed of the side-chains of Asp 210, Asn 212 and His 309 (Figure 4D).

Figure 4: Predicted 3D structure of yak APEX1 and superimposed 3D structure of yak and human APEX1. A: predicted 3D structure of Tianzhu white yak APEX1. B: superimposed 3D structure of yak (red) and human (blue) APEX1. C: The hydrogen bonding network formed by different amino acid residues in yak APEX1, D: Comparison of the binding site residues of Tianzhu white yak (red-rodlike) with human (blue-rodlike). Dotted lines stand for the distance of yak metal ion and active sites (red-globular). The bonding position of metal ion and human active residues were shown by Green-rodlike structure.

Expression of APEX1 gene by Real-time PCR

The mRNA expression levels of APEX1 in the liver were taken as a reference sample and the expression of GAPDH as a house keeping gene. The relative expression levels of APEX1 in the heart, testis, ovary, lung, brain, muscle, spleen and kidney was compared with that of the liver. The expression of APEX1 was detected from all tissues, and the mRNA expression level in brain was highest, then followed by ovary, spleen and kidney (Figure 5).

Figure 5: The mRNA expression level of APEX1 from different tissues in Tianzhu white yak, GAPDH gene was served as the internal reference.

Yak has special physiological and anatomical characteristics, enabling them to live in the very frigid and hypoxemia climates under direct exposure to burning sunlight and natural UV radiations, which may attribute to gene mutation or deletion. Cellular response to oxidative stress is a highly regulated and complex biological process [25], however, no researches have been performed on yak to examine how this animal can overcome the deleterious effect produced by direct sun exposure and how it can repair probable DNA and other lesions induced by different DNA-damaging agents and what kind of mechanisms in adaption of Qinghai-Tibetan Plateau. APEX1 is a dual function protein involved both in the BER pathways of DNA lesions, acting as the major apurinic/apyrimidinic endonuclease, and in eukaryotic transcriptional regulation of gene expression [26]. APEX1 functional activation is a consequence of different stimuli that may generate both physiological and toxic oxidative stress conditions or increase the intracellular cAMP levels leading to different outcomes [27]. Therefore, we supposed that APEX1 may play a key role in high altitude adaption and production performance in yak.

In this study, APEX1 nucleotide sequence was cloned for the first time from Tianzhu white yak, and a full-length cDNA was obtained. The results suggested that the gene sequence of yak and cattle have a high homology (94.4%), which encoded 318 amino acid residues and highly similar to other mammalians (318 amino acids) with the predicted molecular weight 35.1KDa and pI 7.82, which similar to Camelus dromedaries APEX1 [28], but the 193^rd and 202^nd amino acid in yak APEX1 was aminothiopropionic acid (Cys), which were different from aspartic acid (Asp) in other mammalians. The 219^th amino acid was aminopropionic acid (Ala), which different from aspartic acid (Asp) in other mammalians. According to the yak APEX1 amino acid analysis, it was a kind of hydrophilic protein, but not secreted and transmembrane protein, which means that APEX1 protein was not secreted by some organelles, maybe it was transported to DNA damage position for DNA base repair by blood transportation.

The amino acid alignment of the APEX1 in yak and seven other mammalian species have shown that the C-terminus was more conserved than the N-terminus. Despite of this finding, it has been reported that the N-terminal domain may play a role in the fine regulation of the AP endonuclease activity of APEX1. Among the eight highly conserved lysine residues located in site K3, K4, K6, K7, K24, K25, K27, K31, K32 and K35, only five of which (K24, K25, K27, K31 and K32) were involved in the interaction of APEX1 with both RNA and NPM1 [29]. The comparison between the predicted amino acid sequence of APEX1 and the sequences of conserved domains from different organisms indicated that this protein belonged to Human Ape1-like subfamily (EEP family, cd 09087). Compared with other members in EEP family, APEX1 had a His residue while the others kept Tyr [30]. The catalysis sequence contained two divalent metal ion biding sites for Mg²⁺ and Mn²⁺ (at 68, 96, 210, 212, 308 and 309 sites) [31]. Amino acid sequence comparison with Methanothermobacter thermautotrophicus Mth 212 also indicated highly conserved His-Asn- His residues in APEX1 in place of Tyr-Asn-His in EEP family [32].

The real-time PCR results revealed that the highest mRNA expression level of yak APEX1 was in the brain among all tissue samples from the slaughtered animals, and then the expression of APEX1 gene in ovary and spleen were higher than that in heart, testis, lung, muscle and kidney. The higher expression level was expected as APEX1 and other DNA repair machinery was important to correct mistakes and oxidized bases in DNA of the highly dividing cells, like in brain. The brain is an additional extragondal site of estrogen synthesis, and it can synthesize estrogen from cholesterol [33,34], which means brain and ovary are the pivotal organ for control the APEX1 gene or protein synthesis.

The research was supported by the grant from Chinese National 863 plan project (Project No.2013AA102505-3). we are grateful for the worker of Slaughterhouse in Tianzhu county (Gansu Province, China) for their assistance in yak samples collection.