Alternative polyadenylation (APA) is an evolutionarily conserved mechanism for regulating gene expression. Transcript 3' end shortening through changes in polyadenylation site usage occurs following T cell activation, but the consequences of APA on gene expression are poorly understood. We previously showed that GU-rich elements (GREs) found in the 3' untranslated regions of select transcripts mediate rapid mRNA decay by recruiting the protein CELF1/CUGBP1. Using a global RNA sequencing approach, we found that a network of CELF1 target transcripts involved in cell division underwent preferential 3' end shortening via APA following T cell activation, resulting in decreased inclusion of CELF1 binding sites and increased transcript expression. We present a model whereby CELF1 regulates APA site selection following T cell activation through reversible binding to nearby GRE sequences. These findings provide insight into the role of APA in controlling cellular proliferation during biological processes such as development, oncogenesis and T cell activation.

Recent genome-wide analyses have implicated alternative polyadenylation - the process of regulated mRNA 3' end formation - as a critical mechanism that influences multiple steps of mRNA metabolism in addition to increasing the protein-coding capacity of the genome. Although the functional consequences of alternative polyadenylation are well known, protein factors that regulate this process are poorly characterized. Previously, we described an evolutionarily conserved family of neuronal splice variants of the CstF-64 mRNA, βCstF-64, that we hypothesized to function in alternative polyadenylation in the nervous system. In the present study, we show that βCstF-64 mRNA and protein expression increase in response to nerve growth factor (NGF), concomitant with differentiation of adrenal PC-12 cells into a neuronal phenotype, suggesting a role for βCstF-64 in neuronal gene expression. Using PC-12 cells as model, we show that βCstF-64 is a bona fide polyadenylation protein, as evidenced by its association with the CstF complex, and by its ability to stimulate polyadenylation of luciferase reporter mRNA. Using luciferase assays, we show that βCstF-64 stimulates polyadenylation equivalently at the two weak poly(A) sites of the β-adducin mRNA. Notably, we demonstrate that the activity of βCstF-64 is less than CstF-64 on a strong polyadenylation signal, suggesting polyadenylation site-specific differences in the activity of the βCstF-64 protein. Our data address the polyadenylation functions of βCstF-64 for the first time, and provide initial insights into the mechanism of alternative poly(A) site selection in the nervous system.

Two core polyadenylation elements (CPE) located in the 3' untranslated region of eukaryotic pre-mRNAs play an essential role in their processing: the polyadenylation signal (PAS) AAUAAA and the cleavage site (CS), preferentially a CA dinucleotide. Herein, we characterized PAS and CS sequences in a set of cancer predisposition genes (CPGs) and performed an in silico investigation of microRNAs (miRNAs) regulation to identify potential tumor-suppressive and oncogenic miRNAs. NCBI and alternative polyadenylation databases were queried to characterize CPE sequences in 117 CPGs, including 81 and 17 known tumor suppressor genes and oncogenes, respectively. miRNA-mediated regulation analysis was performed using predicted and validated data sources. Based on NCBI analyses, we did not find an established PAS in 21 CPGs, and verified that the majority of PAS already described (74.4%) had the canonical sequence AAUAAA. Interestingly, "AA" dinucleotide was the most common CS (37.5%) associated with this set of genes. Approximately 90% of CPGs exhibited evidence of alternative polyadenylation (more than one functional PAS). Finally, the mir-192 family was significantly overrepresented as regulator of tumor suppressor genes (P < 0.01), which suggests a potential oncogenic function. Overall, this study provides a landscape of CPE in CPGs, which might be useful in development of future molecular analyses covering these frequently neglected regulatory sequences.

VASA is a member of the DEAD-box protein family that plays an indispensable role in mammalian spermatogenesis, particularly during meiosis. In the present study, we isolated, sequenced, and characterized VASA gene in buffalo testis. Here, we demonstrated that VASA mRNA is expressed as multiple isoforms and uses four alternative transcriptional start sites (TSSs) and four different polyadenylation sites. The TSSs identified by 5'-RNA ligase-mediated rapid amplification of cDNA ends (RLM-5'-RACE) were positioned at 48, 53, 85, and 88 nucleotides upstream relative to the translation initiation codon. 3'-RACE experiment revealed the presence of tandem polyadenylation signals, which lead to the expression of at least four different 3'-untranslated regions (209, 233, 239 and 605 nucleotides). The full-length coding region of VASA was 2190 bp, which encodes a 729 amino acid (aa) protein containing nine consensus regions of the DEAD box protein family. VASA variants are highly expressed in testis of adult buffalo. We found five variants, one full length VASA (729 aa) and four splice variants VASA 2, 4, 5, 6 (683, 685, 679, 703 aa). The expression level of VASA 1 was significantly higher than rest of all (P < 0.05) except VASA 6. The relative ratio for VASA 1:2:4:5:6 was 100:1.0:1.6:0.9:48.

To clarify the divergence of the growth hormone receptor (GHR) family, we characterized a novel GHR from a teleost fish (rainbow trout). A 2357-nt cDNA was isolated and found to contain a single initiation site 71 nt from the most 5' end, an open reading frame of 1971 nt encoding a 657-amino acid protein, and a single polyadenylation site 229 nt from the poly-A tail. Based on structural analysis, the protein was identified as a type 1 GHR (GHR1). The new GHR1 shares 42% and 43% amino acid identity, respectively, with GHR2a and GHR2b, the two type 2 GHRs isolated from trout previously. GHR1 mRNA was found in a wide array of tissues with the highest expression in the liver, red muscle, and white muscle. Fasting animals for 4 weeks reduced steady state levels of GHR1 in the liver, adipose, and red muscle. These findings help clarify the divergence and nomenclature of GHRs and provide insight into the function of duplicated GHR types.

RNA sequencing (RNA-Seq) allows for the identification of novel exon-exon junctions and quantification of gene expression levels. We show that from RNA-Seq data one may also detect utilization of alternative polyadenylation (APA) in 3' untranslated regions (3' UTRs) known to play a critical role in the regulation of mRNA stability, cellular localization and translation efficiency. Given the dynamic nature of APA, it is desirable to examine the APA on a sample by sample basis. We used a Poisson hidden Markov model (PHMM) of RNA-Seq data to identify potential APA in human liver and brain cortex tissues leading to shortened 3' UTRs. Over three hundred transcripts with shortened 3' UTRs were detected with sensitivity >75% and specificity >60%. Tissue-specific 3' UTR shortening was observed for 32 genes with a q-value ≤ 0.1. When compared to alternative isoforms detected by Cufflinks or MISO, our PHMM method agreed on over 100 transcripts with shortened 3' UTRs. Given the increasing usage of RNA-Seq for gene expression profiling, using PHMM to investigate sample-specific 3' UTR shortening could be an added benefit from this emerging technology.

It is widely known that transcriptional diversity contributes greatly to biological regulation in eukaryotes. With the development of next-generation sequencing (NGS) technologies, several studies on RNA sequencing have considerably improved our understanding of transcriptome complexity. However, obtaining full-length (FL) transcripts remains a considerable challenge because of difficulties in short read-based assembly. In the present study, single-molecule real-time (SMRT) sequencing and NGS were combined to generate the complete and FL transcriptome of Manis javanica. The results provide a comprehensive set of reference transcripts and hence contribute to the improved annotation of the M. javanica genome. We obtained 45,530 high-confidence transcripts from 19,109 genic loci, of which 8014 genes have not yet been annotated within the M. javanica genome. Furthermore, we revealed 8824 long-chain noncoding RNAs (lncRNAs). A total of 30,199 alternative splicing (AS) and 11,184 alternative polyadenylation (APA) events were identified in the sequencing data. The structure and expression level of 59 digestive enzyme genes, including 13 carbohydrase genes, 28 lipase genes and 18 protease genes, were analyzed, which might provide original data for further research on M. javanica.

Catalase-peroxidases represent one important subfamily of ancestral antioxidant enzymes originally evolved in bacteria for the protection against various forms of oxidative stress. KatG genes coding for these bifunctional catalase-peroxidases were during their peculiar evolution transferred from Bacteroidetes to the fungal phylum Ascomycota via a horizontal gene transfer event. Here we analyse a newly discovered fungal katG gene without introns coding for a thermostable catalase-peroxidase from Chaetomium thermophilum var. dissitum and compare it with closely related thermophilic and mesophilic katGs and their translation products. We show that CthediskatG gene resembling its bacterial counterparts has a typical eukaryotic transcription start site and also contains a conserved eukaryotic polyadenylation signal behind its 3' terminus. Moreover, we have detected polyA tails in corresponding transcripts of katG from two different mRNA libraries of C. thermophilum var. disstum. Although otherwise highly conserved, only in katG genes of two C. thermophilum variants a unique 60 bp long deletion leading in the translated product with high probability to a modified loop and thus access to the prosthetic heme group was observed. We also present an updated molecular phylogeny revealing the evolutionary position of fungal thermostable catalase-peroxidases within a robust phylogenetic tree of the whole KatG subfamily.

The human genome project and subsequent gene annotation projects have shown that the human genome contains 22,000-25,000 functional genes. Therefore, it is believed that the diversity of protein repertoire is achieved by the alternative splicing (AS) mechanism. Transposable elements (TEs) are mobile in nature and can therefore alter their position in the genome. The insertion of TEs into a new gene region can result in AS of a particular transcript through various mechanisms, including intron retention, and alternative donor or acceptor splice sites. TE-derived AS is thought to have played a part in primate evolution and in hominid radiation. However, TE-derived AS or genetic instability may sometimes result in genetic disorders. For the past two decades, numerous studies have been performed on TEs and their role in genomes. Accumulating evidence shows that the term 'junk DNA', previously used for TEs is a misnomer. Recent research has indicated that TEs may have clinical potential. However, to explore the feasibility of using TEs in clinical practice, additional studies are required. This review summarizes the available literature on TE-derived AS, alternative promoter, and alternative polyadenylation. The review covers the effects of TEs on coding genes and their clinical implications, and provides our perspectives and directions for future research.

The Mas protooncogene encodes a G protein-coupled receptor with the common seven transmembrane domains, expressed mainly in the testis and brain. We provided evidence that Mas is a functional angiotensin-(1-7) receptor and can interact with the angiotensin II type 1 (AT1) receptor. The gene is transcriptionally regulated during development in the brain and testis, but its structure was unresolved. In this study we used 5'- and 3'-RACE, RT-PCR, and RNase-protection assays to elucidate the complete Mas gene structure and organization. We identified 12 exons in the mouse Mas gene with 11 in the 5' untranslated mRNA, which can be alternatively spliced. We also showed that Mas transcription can start from 4 tissue-specific promoters, whereby testis-specific Mas mRNA is transcribed from two upstream promoters, and the expression of Mas in the brain starts from two downstream promoters. Alternative splicing and multiple promoter usage result in at least 12 Mas transcripts in which different 5' untranslated regions are fused to a common coding sequence. Moreover, termination of Mas mRNA is regulated by two different polyadenylation signals. The gene spans approximately 27 kb, and the longest detected mRNA contains 2,451 bp. Thus, our results characterize the Mas protooncogene as the gene with the most complex gene structure of all described members of the gene family coding for G protein-coupled receptors.

In this study, a pyruvate carboxylase gene (PYC) from a marine fungus Penicillium viticola 152 isolated from marine algae was cloned and characterized by using Genome Walking method. An open reading frame (ORF) of The PYC gene (accession number: KM593097) had 3582bp encoding 1193 amino acid protein (isoelectric point: 5.01) with a calculated molecular weight of 131.2757kDa. A putative promoter (intronless) of the gene was located at -666bp and contained a TATA box, several CAAT boxes, the 5'-SYGGRG-3' and a 5'-HGATAR-3' sequences. A consensus polyadenylation site (AATAAA) was also observed at +10bp downstream of the ORF. The protein deduced from the PYC gene had no signal peptide, was a homotetramer (4), and had the four functional domains. Furthermore, PYC protein also had three potential N-linked glycosylation sites, among them, -N-S-T-I- at 36 amino acid, -N-G-T-V- at 237 amino acid, and -N-G-S-S- at 517 amino acid were the most possible N-glycosylation sites. After expression of the PYC gene of P. viticola 152 in medium supplemented with CSL and biotin, it was found that the specific pyruvate carboxylase activity in MA production medium supplemented with CSL was much higher (0.5U/mg) than in MA medium supplemented with biotin (0.3U/mg), suggesting that optimal concentration of CSL is required for increased expression of the PYC gene, which is responsible for high level production of malic acid in P. viticola 152 strain.

African clawed frogs, Xenopus laevis, although primarily aquatic, have a high tolerance for dehydration, being capable of withstanding the loss of up to 32-35% of total water body water. Recent studies have shown that microRNAs play a role in the response to dehydration by the liver, kidney and ventral skin of X. laevis. MicroRNAs act by modulating the expression of mRNA transcripts, thereby affecting diverse biochemical pathways. In this study, 43 microRNAs were assessed in frog brains comparing control and dehydrated (31.2±0.83% of total body water lost) conditions. MicroRNAs of interest were measured using a modified protocol which employs polyadenylation of microRNAs prior to reverse transcription and qPCR. Twelve microRNAs that showed a significant decrease in expression (to 41-77% of control levels) in brains from dehydrated frogs (xla-miR-15a, -150, -181a, -191, -211, -218, -219b, -30c, -30e, -31, -34a, and -34b) were identified. Genomic analysis showed that the sequences of these dehydration-responsive microRNAs were highly conserved as compared with the comparable microRNAs of mice (91-100%). Suppression of these microRNAs implies that translation of the mRNA transcripts under their control could be enhanced in response to dehydration. Bioinformatic analysis using the DIANA miRPath program (v.2.0) predicted the top two KEGG pathways that these microRNAs collectively regulate: 1. Axon guidance, and 2. Long-term potentiation. Previous studies indicated that suppression of these microRNAs promotes neuroprotective pathways by increasing the expression of brain-derived neurotrophic factor and activating anti-apoptotic pathways. This suggests that similar actions may be triggered in X. laevis brains as a protective response to dehydration.

The complete mitochondrial genome of the assassin bug Brontostoma colossus (Distant, 1902) (Heteroptera: Reduviidae) has been sequenced using a genome-skimming approach on an Illumina Hiseq 2000 platform. Fifty-four additional heteropteran mitogenomes, including five assassin bug species, were retrieved to allow for comparisons and phylogenetic analyses. The mitochondrial genome of B. colossus was determined to be 16,625 bp long, and consists of 13 protein-coding genes (PCGs), 23 transfer-RNA genes (tRNAs), two ribosomal-RNA genes (rRNAs), and one control region. The nucleotide composition is biased toward adenine and thymine (A+T=73.4%). Overall, architecture, nucleotide composition and genome asymmetry are similar among all available assassin bug mitogenomes. All PCGs have usual start-codons (Met and Ile). Three T and two TA incomplete termination codons were identified adjacent to tRNAs, which was consistent with the punctuation model for primary transcripts processing followed by 3' polyadenylation of mature mRNA. All tRNAs exhibit the classic clover-leaf secondary structure except for tRNASer(AGN) in which the DHU arm forms a simple loop. Two notable features are present in the B. colossus mitogenome: (i) a 131 bp duplicated unit including the complete tRNAArg gene, resulting in 23 potentially functional tRNAs in total, and (ii) a 857 bp duplicated region comprising 277 bp of the srRNA gene and 580 bp of the control region. A phylogenetic analysis based on 55 true bug mitogenomes confirmed that B. colossus belongs to Reduviidae, but contradicted a widely accepted hypothesis. This highlights the limits of phylogenetic analyses based on mitochondrial data only.

Ginkgo biloba L. is regarded as the most ancient living tree, and its kernel has been used as a traditional Chinese medicine for more than 2,000 years. The leaf extracts of this tree have been among the bestselling herbal remedies in Western countries since the last century. To understand the biosynthesis of the pharmacologically active ingredients in G. biloba, flavonoids and terpenoid trilactones (TTLs), we sequenced the transcriptomes of G. biloba leaves, kernels and testae with Iso-Seq and RNA-Seq technologies and obtained 152,524 clean consensus reads. When these reads were used to improve the annotation of the G. biloba genome, 4,856 novel genes, 25,583 new isoforms of previously annotated genes and 4,363 lncRNAs were discovered. Gene ontology and Kyoto Encyclopedia of Genes and Genomes analyses indicated that genes involved in growth, regulation and response to stress were more likely to be regulated by alternative splicing (AS) or alternative polyadenylation (APA), which represent the two most important posttranscriptional regulation mechanisms. It was found that some of the characterized genes involved in the biosynthesis of flavonoids and TTLs were also possibly regulated by AS and APA. Using phylogenetic and gene expression pattern analyses, some candidate genes for the biosynthesis of flavonoids and TTLs were screened. After qRT-PCR validation, the final candidate genes for flavonoid biosynthesis included three UDP-glycosyltransferases and one MYB transcription factor, while the candidate genes for TTL biosynthesis included two cytochrome P450 and one WRKY transcription factor. Our study suggested that Iso-Seq may play an important role in improving genome annotation, elucidating AS and APA mechanisms and discovering candidate genes involved in the biosynthesis of some secondary metabolites.

The crustacean hyperglycemic hormone (CHH) family is an important group of neuropeptides involved in controlling growth, reproduction, and stress response in decapod species. In this study, a new gene containing 4 exons-3 introns flanked by canonical 5'-GT-AG-3' intron splice-site junctions was isolated from Litopenaeus vannamei. Two full length transcripts of this CHH were isolated from eyestalk and pericardial tissue of males and females using rapid amplification of cDNA ends (RACE). Transcripts sequences were 1578bp in length in males pericardial tissues and in males and females eyestalk with 100% identity, but the transcript isolated from females pericardial tissues was shorter (974bp). The differences in transcripts length is a result of two polyadenylation sites present in the 3'UTR resulting in two transcription termination signals. Transcript sequences encoded one unique protein that can be classified as type I CHH subfamily because of the 4 exons and 3 introns structure, although the CPRP region is not-well conserved and there is no amidation in the C-terminal of the deduced amino acid sequence. Furthermore, there is a glycine inserted in the mature peptide not at position 12 as in type II CHHs but after amino acid 31 and the phylogenetic analysis did not group the peptide within type I, but closer to type II CHHs. We demonstrated by endpoint-PCR, qPCR, and in situ hybridization (ISH), that this gene is expressed in neuroendocrine organs known to express CHHs in penaeid shrimp, including X-organ and optic nerve in eyestalk, supraesophageal ganglion (SoG), but it is also expressed in other organs as gill, gut, pericardial cavity, as well as in terminal ampoule or spermatophore and vas deferens of males.

FUS, an RNA-binding protein (RBP), is mutated or abnormally regulated in neurodegenerative disorders. FUS regulates various aspects of RNA metabolisms. FUS-binding sites are rich in GU contents and are highly degenerative. FUS-binding motifs of GGU, GGUG, GUGGU and CGCGC have been previously reported. These motifs, however, are applicable to a small fraction of FUS-binding sites. As CLIP-seq tags are enriched in genes that are highly expressed, we normalized CLIP-seq tags by Nascent-seq tags or RNA-seq tags of mouse N2a cells. Nascent-seq identifies nascent transcripts before being processed for splicing and polyadenylation. We extracted frequently observed 4-nt motifs from Nascent-seq-normalized CLIP regions, RNA-seq-normalized CLIP regions, and native CLIP regions. Specific GU-rich motifs were best detected in Nascent-seq-normalized CLIP regions. Analysis of structural motifs using Nascent-seq-normalized CLIP regions also predicted GU-rich sequence forming a stem structure. Sensitivity and specificity were calculated by examining whether the extracted motifs were present at the cross-linking-induced mutation sites (CIMS), where FUS was directly bound. We found that a combination of six motifs (UGUG, CUGG, UGGU, GCUG, GUGG, and UUGG), which were extracted from Nascent-seq-normalized CLIP-regions, had a better discriminative power than (i) motifs extracted from RNA-seq-normalized CLIP regions, (ii) motifs extracted from native CLIP regions, (iii) previously reported individual motifs, or (iv) 15 motifs in SpliceAid 2. Validation of the 6 GU-rich (6GUR) motifs using CLIP-seq of the cerebrum and the whole brain showed that the 6GUR motifs were specifically enriched in CIMS. The number of the 6GUR motifs in an uninterrupted region was counted and multiplied by four to calculate the area, which was defined as the 6GUR-Score. The 6GUR-Score of 8 or more best discriminated CIMS from CIMS-flanking regions. We propose that the 6GUR motifs predict FUS-binding sites more efficiently than previously reported individual motifs or 15 motifs in SpliceAid 2.

The PVGOX gene (accession number: KT452630) was isolated from genomic DNA of the marine fungi Penicillium viticola F1 by Genome Walking and their expression analysis was done by Fluorescent RT-PCR. An open reading frame of 1806bp encoding a 601 amino acid protein (isoelectric point: 5.01) with a calculated molecular weight of 65,535.4 was characterized. The deduced protein showed 75%, 71%, 69% and 64% identity to those deduced from the glucose oxidase (GOX) genes from different fungal strains including; Talaromyces variabilis, Beauveria bassiana, Aspergillus terreus, and Aspergillus niger, respectively. The promoter of the gene (intronless) had two TATA boxes around the base pair number -88 and -94 and as well as a CAAT box at -100. However, the terminator of the PVGOX gene does not contain any polyadenylation site (AATAAA). The protein deduced from the PVGOX gene had a signal peptide containing 17 amino acids, three cysteine residues and six potential N-linked glycosylation sites, among them, -N-K-T-Y- at 41 amino acid, -N-R-S-L- at 113 amino acid, -N-G-T-I- at 192 amino acid, -N-T-T-A at 215 amino acid, -N-F-T-E at 373 amino acid and -N-V-T-A- at 408 amino acid were the most possible N-glycosylation sites. Furthermore, the relative transcription level of the PVGOX gene was also stimulated in the presence of 4% (w/v) of calcium carbonate and 0.5 % (v/v) of CSL in the production medium compared with that of the PVGOX gene when the fungal strain F1 was grown in the absence of calcium carbonate and CSL in the production medium, suggesting that under the optimal conditions, the expression of the PVGOX gene responsible for gluconic acid biosynthesis was enhanced, leading to increased gluconic acid production. Therefore, the highly glycosylated oxidase enzyme produced by P. viticola F1 strain might be a good producer in the fermentation process for the industrial level production of gluconic acid.

Bidirectional promoters are defined as those that regulate adjacent genes organized in a divergent fashion (head to head orientation) and separated by <1 kb. In order to dissect bidirectional promoter activity in a model plant, deletion analysis was performed for seven rice promoters using promoter-reporter gene constructs, which identified three promoters to be bidirectional. Regulatory elements located in or close to the 5'-untranslated regions (UTR) of one of the genes (divergent gene pair) were found to be responsible for their bidirectional activity. DNA footprinting analysis identified unique protein binding sites in these promoters. Deletion/alteration of these motifs resulted in significant loss of expression of the reporter genes on either side of the promoter. Changes in the motifs at both the positions resulted in a remarkable decrease in bidirectional activity of the reporter genes flanking the promoter. Based on our results, we propose a novel mechanism for the bidirectionality of rice bidirectional promoters.

Microphthalmia-associated transcription factor (MITF) is a basic helix-loop-helix leucine zipper (bHLH-LZ) transcription factor, which regulates the differentiation and development of melanocytes and pigment cell-specific transcription of the melanogenesis enzyme genes. Though multiple splice variants of MITF have been reported in humans, mice and other vertebrate species, in merino sheep (Ovis aries), MITF gene splicing has not yet been investigated until now. To investigate the sheep MITF isoforms, the full length mRNA/cDNAs from the skin of merino sheep were cloned, sequenced and characterized. Reverse transcriptase (RT)-PCR analysis and molecular prediction revealed two basic splice variants with (+) and without (-) an 18 bp insertion viz. CGTGTATTTTCCCCACAG, in the coding region (CDS) for the amino acids 'ACIFPT'. It was further confirmed by the complete nucleotide sequencing of splice junction covering intron-6 (2463 bp), wherein an 18bp intronic sequence is retained into the CDS of MITF (+) isoform. Further, full-length cDNA libraries were enriched by the method of 5' and 3' rapid amplification of cDNA ends (RACE-PCR). A total of seven sheep MITF splice variants, with distinct N-terminus sequences such as MITF-A, B, E, H, and M, the counterparts of human and mouse MITF, were identified by 5' RACE. The other two 5' RACE products were found to be novel splice variants of MITF and represented as 'MITF truncated form (Trn)-1, 2'. These alternative splice (AS) variants were illustrated using comparative genome analysis. By means of 3' RACE three different MITF 3' UTRs (625, 1083, 3167bp) were identified and characterized. We also demonstrated that the MITF gene expression determined at transcript level is mediated via an intron-6 splicing event. Here we summarize for the first time, the expression of seven MITF splice variants with three distinct 3' UTRs in the skin of merino sheep. Our data refine the structure of the MITF gene in sheep beyond what was previously known in humans, mice, dogs and other mammals.

Pyridoxal kinase is a key enzyme for the biosynthesis of pyridoxal 5'-phosphate. Pyridoxal 5'-phosphate is the catalytically active form of vitamin B6, and acts as a cofactor in >140 different enzyme reactions. It is still unknown how the kinase synthesis is regulated in the cells, and nothing has been reported about the gene promoter. In the present study, based on the bioinformatics analysis of the 5'-flanking region of the human PDXK gene, we cloned the promoter region by PCR. Through the construction of a series of luciferase expression vectors containing the human PDXK promoter region, we characterized the promoter in terms of its structure and function. The transcription start site is at 198bp upstream of the ATG translation initiation site. An important regulatory region is located at -665/-433bp upstream of the transcription start site. The promoter lacks the canonical TATA box, but contains three GC-boxes and one E-box. A deletion and mutation experiment revealed that the transcription factor Sp1 binding site C (-553/-543) is critical in maintaining the robust promoter activity. Knockdown of Sp1 by RNA interference and chromatin immunoprecipitation analysis further proved that the Sp1 is involved in the regulation of the PDXK gene expression.