Bioinformatic and expression analysis of novel porcine β-defensins

Bioinformatic and expression analysis of novel porcine β-defensins

Please download to get full document.

View again

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.


Publish on:

Views: 2 | Pages: 8

Extension: PDF | Download: 0

  Bioinformatic and expression analysis of novel porcine b -defensins Yongming Sang, 1 Amar A. Patil, 2 Guolong Zhang, 2 Chris R. Ross, 1 Frank Blecha 1 1 Department of Anatomy and Physiology, 228 Coles Hall, Kansas State University, Manhattan, Kansas 66506, USA 2 Department of Animal Science, Oklahoma State University, Stillwater, Oklahoma 74078, USA Received: 14 November 2005 / Accepted: 5 January 2006 Abstract  b -Defensins are a major group of mammalian anti-microbial peptides. Although more than 30  b -de-fensins have been identified in humans, only oneporcine  b -defensin has been reported. In this articlewe report the identification and initial character-ization of 11 novel porcine  b -defensins (pBD). Usingbioinformatic approaches, we screened 287,821 por-cine expressed sequence tags for similarity of theirpredicted peptides to known human  b -defensins andidentified full-length or partial sequences for the 11novel pBDs. Similar to the previously identifiedpBD1, all of these peptides have a consensus  b -de-fensin motif. A differential expression pattern forthese newly identified genes was found. For exam-ple, unlike most  b -defensins,  pBD2  and  pBD3  wereexpressed in bone marrow and in other lymphoidtissues including thymus, spleen, lymph nodes,duodenum, and liver. Including  pBD2  and  pBD3 , sixporcine  b -defensins were expressed in lung and skin.Several newly identified porcine  b -defensins,including  pBD123 ,  pBD125 , and  pBD129 , were ex-pressed in male reproductive tissues, including lob-uli testis and some segments of the epididymis.Phylogenetic analysis indicates that in most casesthe evolutionary relationship between individualporcine  b -defensins and their human orthologs iscloser than the relationship among  b -defensins inthe same species. These findings establish the exis-tence of multiple porcine  b -defensins and suggestthat the pig may be an ideal model for the charac-terization of  b -defensin diversity and function. Introduction  Antimicrobial peptides are small cationic polypep-tides that function as one of the earliest mediators ofhost defense in many species of insects, plants, andanimals (Boman 1995, 2003). In humans and other mammals, defensins are a major family of antimi-crobial peptides whose two main subfamilies,  a  and b , are characterized by  b -sheet folds and a frameworkof six disulfide-linked cysteines (Ganz 2003; Lehrerand Ganz 2002; Schutte and McCray 2002). The  a -and  b -defensins differ in their cysteine disulfidepairings and in the number of amino acids betweenthe six cysteines. For  b -defensins, the six cysteinesare linked at positions 1-5, 2-4, and 3-6. The canon-ical sequence represented by human  b -defensinsusually is X 2-10 -C-X 6 -C-X 3-4 -C-X 9-13 -C-X 4-7 —C-CX n ,where X represents any amino acid residue (Ganz2003; Patil et al. 2005; Selsted and Ouellette 2005).The third defensin subfamily,  h -defensins, is struc-turally unlike  a - and  b -defensins and is expressed inOld World monkeys and orangutans (Nguyen et al.2003; Selsted and Ouellette 2005).At least one and usually several defensins havebeen identified in all mammals that have beenstudied (Boman 2003; Ganz 2003). However, tissue distribution profiles and existence of subfamily de-fensins vary greatly even between closely relatedspecies. For example, humans have  a -defensins inleukocytes and intestinal Paneth cells, and  b -defen-sins are found in many epithelial cells. Mice and ratshave Paneth cell  a -defensins and epithelial cell b -defensins; however, although mice lack leukocyte a -defensins, rats possess neutrophil  a -defensins (Bo-man 2003; Ganz 2003; Patil et al. 2004, 2005; Yang  Abbreviations : DEFB = human  b -defensin; EST = expressed se-quence tag; GAPDH = glyceraldehyde 3-phosphate dehydrogenase;pBD = porcine  b -defensin; RT-PCR = reverse-transcriptase poly-merase chain reaction.The nucleotide sequence data reported in this article have beensubmitted to GenBank.Correspondence to: Frank Blecha; E-mail: 332  DOI: 10.1007/s00335-005-0158-0    Volume 17, 332  339 (2006)      Springer Science+Business Media, Inc. 2006  et al. 2004). Bovine neutrophils have several  b -de-fensins and epithelial  b -defensins are expressed inbovine trachea, tongue, and intestine (Diamond et al.1991; Ganz 2003; Selsted et al. 1993). To date, onlyone epithelial  b -defensin (pBD1) has been identifiedin pigs and defensins have not been detected inporcine leukocytes (Ganz 2003; Zhang et al. 1998, 1999). Moreover, there is no evidence that  a -defen-sins are present in pigs (Ganz 2003).Although considerable progress has been madein identifying complete defensin repertoires in sev-eral species including humans, chimpanzees, mice,rats, dogs, and chickens (Kao et al. 2003; Patil et al.2004, 2005; Rodriguez-Jimenez et al. 2003; Schutte et al. 2002; Xiao et al. 2004), information on thecomplete repertoire of porcine  b -defensins is lacking.Because pigs are often used for comparative physio-logic and immunologic studies and because porcinetissues and organs are often used for xenotransplan-tation, we sought to identify the complete  b -defensinprofile in pigs. In this article we report the identifi-cation and initial characterization of 11 novel por-cine  b -defensins, information that is fundamental tothe comparative investigation of  b -defensins in in-nate immunity. Materials and methods   BLAST-based searches.  The porcine expressed se-quence tag (EST) database was searched withBLASTP and TBLASTN programs (Altschul 1990),using the National Center for Biotechnology Infor-mation (NCBI) website tools (www.ncbi.nlm.nih.-gov/blast/) against the EST collection other thanhuman or mouse (EST_others). Initial queries for thesearch used amino acid sequences for known humandefensins ( DEFB1 - 31 ) (Schutte et al. 2002) and three HE2/EP2  sequences (Frohlich et al. 2000, 2001) as well as some identified bovine  b -defensins (Diamondet al. 1991; Selsted et al. 1993). NCBI defaultparameters were used in the searches and any po-tential hits were curated manually.  EST annotation and defining coding regions.  Mostporcine EST clones represent information definingfull-length  b -defensin cDNA and thus do not needadditional annotation. However, in some cases theESTs were annotated using the stack-PACK version2.2 program ( as described(Lynn et al. 2003, 2004). Briefly, input ESTs weremasked to remove repeat sequences and clustered ifthey shared more than 100 bp at greater than 96 % identity. Representative EST sequences or processedconsensus sequences were used to define the codingregion with ESTScan ( and predicted peptides weretranslated using the Translate program (  Alignment and phylogenetic analysis.  Multiplesequence alignment was performed using the PILE-UP program from the Wisconsin Package Software(Accelrys, San Diego, CA). Amino acid sequencesselected for alignment were three residues before andseveral residues after the six-cysteine motif (Schutteet al. 2002). The comparison matrix was set at Blo-sum 62 with a gap creation penalty of 8 and a gapextension penalty of 2. Phylogenetic and molecularevolutionary analyses were conducted on the mostcomplete peptide sequences using MEGA version 2.1(Kumar et al. 2001).  Expression analysis by semiquantitative RT- PCR and real-time RT-PCR.  Tissues obtained fromhealthy 5-week-old male crossbred pigs, as previ-ously described (Zhang et al. 1998), were chosen torepresent organs of the digestive, pulmonary, andimmune systems and included bone marrow, intes-tine, liver, lung, spleen, thymus, testes, and epidid-ymis. All collection procedures were approved by theKansas State University Institutional Animal Careand Use Committee. Tissue samples were collected,placed immediately in liquid nitrogen, and stored at ) 135  C until use. Total RNA was extracted with TRIRegent (Sigma-Aldrich, St. Louis, MO) after grindingfrozen tissues in liquid nitrogen. A one-step RT-PCRwas used to detect expression of target transcripts.Briefly, total RNA was treated with RQ1 RNase-freeDNAse I (Promega, Madison, WI) to remove possiblegenomicDNAcontamination.RNAsamples(250ng)were run in a 25- l l RT-PCR reaction mixture with a0.1  l M concentration of each sense and antisenseprimer derived from cDNA sequences (Table 1).Semiquantitative RT-PCR was performed using aone-step RT-PCR kit (Qiagen, Valencia, CA). cDNAsynthesis and predenaturation were performed at50  C for 30 min and 95  C for 15 min to activate theantibody-protected DNA polymerase, and amplifi-cation was carried out at 95  C for 30 sec, 55  C for 30sec, and 72  C for 40 sec; final extension was at 72  Cfor 10 min. For most genes, 32, 35, and 40 PCR cycleswere used in different replicates to ensure linearamplification and optimal estimation of relativeexpression levels. After amplification, 10  l l of eachreaction mixture were analyzed by 2 %  agarose gelelectrophoresis, bands were then visualized by ethi-dium bromide staining in a FluorChem TM digitalimaging system (Alpha Innotech Corp., San Leandro,CA), and integrated density values were measuredusing the digital imaging system. Integrated density Y. S ANG ET AL .: N OVEL  P ORCINE  b -D EFENSINS  333  values were standardized with values of the house-keeping gene, glyceraldehyde 3-phosphate dehydro-genase ( GAPDH  ), and presented as a ratio relative tothe expression of  GAPDH  .For genes reported to show expression patternssimilar to their human orthologs, we confirmed theirexpression profiles in tissues using a SYBR-Green-based real-time RT-PCR system (Qiagen). In brief,real-time quantitative RT-PCR was performed on aSmartCycler (Cepheid, Sunnyvale, CA) as previouslydescribed (Sang et al. 2005) with gene-specificprimers as described. DNase-treated total RNA (200ng) was used in each 25- l l RT-PCR reaction. TheRT-PCR cycling conditions were 30 min at 50  C,95  C for 15 min, followed by 45 cycles of 15 sec at95  C, 30 sec at 56  C, and 30 sec at 72  C with theoptic on to perform fluorescence data collection.Amplicon authenticity was confirmed by sequenc-ing before performing real-time RT-PCR. Thresholdcycle (C t ) was determined by exponential productamplification and subsequent increased fluorescenceintensity above background. Relative gene-expres-sion data were normalized against the C t  values ofthe housekeeping gene ( GAPDH  ) and the relative Table 1.  Primer sequences for RT-PCR and real-time RT-PCR analysis  pBDs Primer sequence (5 ¢  to 3 ¢  ) GenBank accession number  a Location in cDNA (nt) pBD2 AY506573 (AW785442)sense ATGAGGGCCCTCTGCTTGCT 53  72antisense ATACTTCACTTGGCCTGTGTGTCC 312  289pBD3 AY460575 (CF789126)sense CTTCCTATCCAGTCTCAGTGTTCTGC 200  225antisense GGCTTCTGTAGACTTCAAGGAGACAT 508  483pBD4 AY460576 (BX672669)sense GTGGCTTGGATTTGAGGAGAGAGT 107  130antisense AGTGATACACAGGCCTGGAAGGAT 339  316pBD104 DQ274056 (BX918848)sense TCCTTCCACGTATGGAGGCTTGTT 300  323antisense TTACAATACCTCCGGCAGCGAGAA 632  608sense AAGACTCCTGTTAGCACCCAGCAT 449  472antisense TTACAATACCTCCGGCAGCGAGAA 632  608pBD108 DQ274057 (BX917425)sense GACGATTGTCATTCTTCTGATCCTGG 33  58antisense TAGGTTGACTTGTGGTGCCCGAAA 291  268sense GTGAGAAAGACCAAGGATCATGCAG 124  148antisense TAGGTTGACTTGTGGTGCCCGAAA 291  268pBD114 BK005518 (BX923414)sense TGTACCTTGGTGGATCCTGAACGA 95  118antisense CGCCCTCTGAATGCAGCATATCTT 221  196sense TGTACCTTGGTGGATCCTGAACGA 95  118antisense ATTCCTACACCTCTCTGTACTGGTGC 304  279pBD123 BK005519 (BX915917)sense AGCCATGAAGTGTTGGAGTGCGTT 76  100antisense GTACACAGCACATAGTTGCATCCC 177  153sense GTGCGTTGGGAAGATGCAGAACAA 93  116antisense AACAGGGTAGGGCCAAGAATGAGT 322  298pBD125 BK005520 (BX926653)sense AGCCATGAATCTCCTGCTGACCTT 32  55antisense TGCAGCATGCTCGCTTGTTCATAC 201  178sense GTGACCAAAGCTGGCTGGAATGTT 81  104antisense TCCTGCTCAGTTCCTGTGCTTTCT 370  347pBD129 BK005521 (BX918362)sense CAAAGACCACTGTGCCGTGAATGA 118  141antisense TTGATGCTGGCGAAAGGGTTGGTA 357  334pEP2C BK005522 (BX925543)sense CCCTTTCCAGGAACCTGAACCAAA 184  208antisense TGGCTTGTAGGCTCTGGAGAACAA 388  365pEP2E BK005523 (BX919973)sense TGCCTTATGCAACATGGAACCTGC 295  318antisense AGGTGCTAGAACCACCATTCATCG 445  422sense TCCAGACACTTCCCTATGGCCTTT 12  35antisense GCCTGCAGGTTCCATGTTGCATAA 322  299 a GenBank accession numbers (EST numbers). 334  Y. S ANG ET AL .: N OVEL  P ORCINE  b -D EFENSINS  index (2 ) DD Ct ) was determined in comparison to theaverage expression levels of control samples with theindex defined as 1.000 (Livak and Schmittgen 2001). Results and discussion  The first porcine  b -defensin, pBD1, was identifiedbased on PCR amplification of tissue RNA withprimers generated from bovine lingual  b -defensin(Zhang et al. 1998, 1999). We now know that iden- tification of pBD1 was quite fortuitous, because ofthe 287,821 EST entries generated from varioussources today, no pBD1 EST was identified. In con-trast, using a bioinformatic approach, we haveidentified 11 novel porcine  b -defensins. Each novelporcine defensin is related to at least one ESTclone whose sequence range covers the six-cysteinedefensin motif. Porcine  b -defensin-2 (pBD2) is highlyexpressed in many tissues and, has the most (19) ESTclones identified, with 12 of these covering the en-tire open reading frame in their registered cDNAsequences. Similarly, pBD3, pBD4, and pBD129 haveabundant EST entries in the database; all are repre-sented by five almost identical ESTs. The other eightpBD candidates are represented by one to three ESTclones and cover all or most of the coding regions ofputative porcine  b -defensins. We did not identify anyporcine EST with an  a -defensin signature in theirtranslated frames using a similar strategy.Alignment of the pBD predicted peptides clearlyshows that they possess typical  b -defensin charac-teristics, such as the six-cysteine  b -defensin motif,spacing patterns of the six cysteines, and represen-tative content of positive-charged residues (Fig. 1 Fig. 1.  Multiple sequence alignment of porcine  b -defensin proteins. Amino acid sequences were predicted from cDNAsequences and aligned with minor manipulations to maximize sequence alignment. Conserved residues are shaded andthe six cysteines are also boxed. The consensus sequence shows cysteines (C), positively charged amino acids (+), and otheramino acids if they are represented in more than 50 % of all predicted  b -defensin proteins. Table 2.  Properties of porcine  b -defensins Porcine b -defensinGenBankaccession number  AminoacidsORF verified  a Six-cysteinespacing  pattern  b H+R+K   c Main expression features in healthy tissues pBD1 AF031666 64 Yes 6 4 9 6 9 Airway and oral mucosapBD2 AY506573 69 Yes 6 4 9 6 8 Liver, intestine d , lung, andbone marrow pBD3 AY460575 67 Yes 6 4 9 6 12 Bone marrow, liver, lung, andlymphoid tissuespBD4 AY460576 67 Yes 6 3 9 6 9 Lung and proximal epididymispBD104 DQ274056    No 6 3 9 5 11 Spleen, liver, and testispBD108 DQ274057 73 Yes 6 3 9 5 10 Liver and proximal epididymispBD114 BK005518 69 Yes 6 3 9 6 8 Ileum, spleen, liver, lung, andmale reproductive tissuespBD123 BK005519    Yes 6 3 9 5 8 Ileum, spleen, lung, and malereproductive tissuespBD125 BK005520 147 Yes 7 3 9 5 13 Lung, thymus, and epididymispBD129 BK005521 >173 Yes 6 3 9 4 11 Epididymis, duodenum, jejunum,spleen, lung, and skinpEP2C BK005522 >108 Yes 6 3 9 6 6 Thymus, skin, testis, and somesections of epididymispEP2E BK005523 85 Yes 6 4 9 6 9 Not evident in tissues evaluated a Open reading frames (ORF) in EST sequence were verified by computational predication or sequencing after PCR amplification. b Number of amino acids that separate the cysteine residues (C1-C2, C2-C3, C3-C4, and C4-C5) in the six-cysteine  b -defensin motif. c Number of positively charged residues (H, histidine; R, arginine; K, lysine) in the putative mature  b -defensin peptides. Calculationsconsider positive residues between seven amino acids before the first cysteine (C1) and, at most, seven amino acids after the last cysteine(C6). d The gene was expressed in all three sections (duodenum, jejunum and ileum) of the intestine.Y. S ANG ET AL .: N OVEL  P ORCINE  b -D EFENSINS  335  and Table 2). In addition, the canonical sequence ofporcine  b -defensins is almost identical to that ofhuman  b -defensins. After identifying the novel por-cine defensins, we used them to query the Swiss-Prot database using BLASTP and assigned a tentativename for each porcine defensin as suggested by theHUGO Gene Nomenclature Committee, UniversityCollege, London, UK ( For example, pBD125 is mosthomologous to human DEFB125 and so on. Excep-tions to this naming convention are pBD2, pBD3,and pBD4, which we previously identified experi-mentally, confirmed by bioinformatics searching,and thus were named serially following pBD1 (Zhanget al. 1998). Most porcine  b -defensins show signifi-cant similarities to their human counterparts in bothsequence length and identity.Generally,  b -defensin precursors consist of lessthan 80 amino acid residues, which are encoded bytwo exons (Ganz 2003; Yang et al. 2004). This characteristic is found in pBD1, pBD2, pBD3,pBD4, and pBD114. Exceptions to this characteris-tic are two groups of defensin-like molecules,including some EP2/HP2 gene products (EP2C/D/E)and DEFB25-29. These defensins are longer becauseof the addition of as many as 20-100 amino acidsafter the N-terminal leader sequence or at the C-terminus adjacent to the C5 and C6 residues(Schutte and McCray 2002). Candidates represent-ing these two groups of  b -defensins also have beenidentified in the pig and are represented bypBD125, pBD129, pEP2C, and pEP2E (Table 2).Two other features we compared are related to thesix-cysteine motif. As shown in Table 2, almost allporcine  b -defensins conserve the number of aminoacids between the six cysteine residues and theratio of positive residues, which contributes to thepositive charge of the cationic peptides and relatesto antimicrobial activity (Schutte et al. 2002; Yanget al. 2004).Phylogenetic analysis of representative humanand porcine  b -defensins indicated that they werederived from a common ancestor. This analysisshowed that individual porcine  b -defensins are in thesame sub-branch as their human homolog, exceptpBD1, which is a more diverse isoform than human b -defensins and has no close human homolog. Fur-thermore, similarity among each subclass of  b -de-fensins from different species is higher than thatbetween different subclasses from the same species.The phylogenetic relationship between human andporcine  b -defensins is supported by the high boot-strap values on the branches, which are based onmultiple resampling of the srcinal data. Bootstrapanalysis is the most common method for estimatingthe degree of confidence in the topology of phyloge-netic trees (Fig. 2) (Kumar et al. 2001; Lynn et al.2004).To evaluate tissue expression profiles of pBDsand verify the accuracy of the EST sequences, wedesigned gene-specific primers based on representa-tive EST sequences (Table 1). Each set of primers wasdesigned from diverse regions on different exons tofacilitate specific amplification of target genes and to Fig. 2.  Phylogenetic tree of humanand porcine  b -defensins. Human b -defensins are DEFBs and porcine b -defensins are pBDs. For branchessupported by bootstrap analysiswith the percentage of 1000replications, the percentage isindicated on the branches. The  bar  indicates the  p -distance. GenBankaccession numbers are the sameTable 2. 336  Y. S ANG ET AL .: N OVEL  P ORCINE  b -D EFENSINS
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks