Gene 373 (2006) 44–51 www.elsevier.com/locate/gene

Cloning and expression of II1 from chicken☆ ⁎ Kyle S. MacLea, Hans H. Cheng

United States Department of Agriculture, Agricultural Research Service, Avian Disease and Oncology Laboratory, 3606 East Mount Hope Road, East Lansing, Michigan 48823, USA

Received 21 November 2005; received in revised form 27 December 2005; accepted 28 December 2005 Available online 24 February 2006 Received by M. Batzer

Abstract

Acid of the deoxyribonuclease II (DNase II, EC 3.1.22.1) family have been implicated in the degradation of DNA from apoptotic cell corpses formed in the process of normal mammalian development. Although a predicted DNase II has been detected in the chicken through expressed sequence tag (EST) analysis, to date no homolog of these important has been identified in vivo in any avian species. Here we report the cloning and expression of DNase II from the chicken, Gallus gallus. When expressed, the 363 amino acid glycoprotein is observed to be approximately 45 kDa in size and to exhibit DNA hydrolytic activity at pH 5 consistent with DNase II in other species. Furthermore, chicken DNase II sequence is compared with an identified partial sequence from the zebra finch, Taeniopygia guttata, as well as the previously identified homologs found in the fowlpox and canarypox viruses and the previously cloned mammalian DNases II. Through analysis of its amino acid sequence, comparative gene structure, and conserved synteny, chicken DNase II appears to represent a member of the DNase IIβ subfamily and the apparent lack of a DNase IIα homolog in the chicken has important evolutionary implications for the study of this gene family. Published by Elsevier B.V.

Keywords: DNase II; DNase IIβ; DLAD; Chicken; Fowlpox; Canarypox

1. Introduction have been identified: DNase IIα (Krieser and Eastman, 1998; Baker et al., 1998) and DNase IIβ (also known as DNase II-like DNase II (EC 3.1.22.1) is an with an acidic Acid DNase or DLAD) (Shiokawa and Tanuma, 1999; Krieser pH optimum, which has been reported in and et al., 2001). Though in these studies DNase IIα was secretions from cells of many organisms (MacLea et al., 2003b; ubiquitously expressed in all the tissues examined, DNase Evans and Aguilera, 2003). In humans and rodents, the IIβ has a more restricted pattern of expression that varies by mammalian organisms in which DNase II has been studied the organism. Both family members play a critical role in most extensively, two members of the DNase II family degradation of undigested apoptotic cell DNA that normally accumulates during development (Kawane et al., 2001; Krieser Abbreviations: BLAST, basic local alignment search tool; EST, expressed et al., 2002; Nishimoto et al., 2003). In particular, the activities sequence tag; EGFP, enhanced green fluorescent protein; ECL, enhanced encoded by these enzymes are critical steps in the processes of chemiluminescence; βME, β-mercaptoethanol; SRED, single radial enzyme definitive erythropoiesis (DNase IIα) and lens cell differenti- diffusion; SD, standard deviation; TM, tunicamycin; RACE, rapid amplification ation (DNase IIβ). of cDNA ends. ☆ The use of trade, firm, or corporation names in this publication is for the Acid endonuclease activity has also been reported in the information and convenience of the reader. Such use does not constitute an chicken (Torriglia et al., 2001). However, despite the cloning of official endorsement or approval by the United States Department of Agriculture DNase II family members in several mammalian organisms, no or the Agricultural Research Service of any product or service to the exclusion of DNase II homolog has been cloned or purified from an avian others that may be suitable. species. Although our previous work had identified a likely ⁎ Corresponding author. Tel.: +1 517 337 6758; fax: +1 517 337 6776. E-mail address: [email protected] (H.H. Cheng). chicken DNase II in expressed sequence tag (EST) sequences 1 The nucleotide sequence of chicken DNase II reported in this paper is (MacLea et al., 2003a), it had never been identified in vivo available from GenBank, accession number DQ272298. before. Here we report the first cloning and expression of an

0378-1119/$ - see front matter. Published by Elsevier B.V. doi:10.1016/j.gene.2005.12.019 K.S. MacLea, H.H. Cheng / Gene 373 (2006) 44–51 45 active avian DNase II. Isolated from the chicken, Gallus gallus, the full-length cDNA (including some sequence from the 5′ and the study of this avian DNase II has potentially important 3′ untranslated regions) and clone it into a eukaryotic expres- implications for understanding the evolutionary history of the sion vector (Fig. 1). DNase II , especially in light of the recently To perform the PCR amplifications, custom oligonucleotides completed genome draft sequence of the chicken (Hillier et al., were synthesized, desalted, and lyophilized, by Operon Bio- 2004). technologies (Huntsville, Alabama). As shown in the schematic diagram (Fig. 1), these primers were used: 2. Materials and methods KM-D2B-F2 (5′-CGCTCACTGTGCCACGCCGAGATG-3′) 2.1. Database searches KM-D2B-R2 (5′-GCTGTTTGGGAGGAACAGTCC-3′)

Using human DNase IIα (AAC77366) and fowlpox Cel1/ DNase II (CAA07012) protein sequences as the query sequences for BLAST (Basic Local Alignment Search Tool), tblastn searches were conducted of GenBank at the National Center for Biotechnology Information (NCBI) website (http:// www.ncbi.nlm.nih.gov/). Further searches for expressed se- quence tags (ESTs) were undertaken using the BBSRC ChickEST Database (http://www.chick.umist.ac.uk/), the Uni- versity of Delaware Chick EST Project (http://www.chickest. udel.edu/), and the Songbird Neurogenomics Initiative EST Project (http://titan.biotec.uiuc.edu/songbird/). Later, to look for further homologues from chicken, the cloned chicken DNase II sequence was also employed as the query sequence for searches of each of these databases and the chicken genome draft sequence (http://www.ensembl.org/Gallus_gallus/).

2.2. Sequence analysis and multiple sequence alignment

Sequence analysis and multiple sequence alignments were undertaken using the ClustalW webserver (http://www.ebi.ac. uk/clustalw/)(Chenna et al., 2003). BOXSHADE 3.21 (http:// www.ch.embnet.org/software/BOX_form.html) was used to create and shade the sequence alignment image. To improve clarity in alignments, Adobe Photoshop CS and Adobe Illustrator CS (Adobe Systems, Inc., San Jose, CA) were used to delineate important protein sequence features. Signal peptide predictions used the SignalP 3.0 webserver (http://www.cbs.dtu. dk/services/SignalP/). Gene structure was determined using data from superimposition of the cDNA on the chicken genome draft sequence and illustrated using Adobe Photoshop and Illustrator.

2.3. Cloning of chicken DNase II and plasmid vector construction Fig. 1. Schematic diagram of chicken DNase II cloning and subsequent EST sequences from earlier studies (MacLea et al., 2003a) eukaryotic expression plasmid construction. Briefly, total cellular mRNA from and new database searches (see Section 2.1 above) were line 0 embryos was purified, reverse transcribed, and amplified using specific assembled using Sequencher 4.5 software (Gene Codes, Ann primers (KM-D2B-F2 and -R2) to isolate the chicken DNase II cDNA. This cDNA was introduced into the pCR4 plasmid vector using the TOPO TA kit Arbor, Michigan). These sequence tags contained DNase II (Invitrogen), creating pcD2B-TOPO. A second round of PCR amplification with sequence: BI392355, BG625517, AJ398432, BI066323, different primers (KM-D2B-F1 and -R1) was used to introduce a Kozak BM440623, BM426634, BI064835, CR386219, BU241744, consensus sequence, 3′ Gly–Ala–Gly linker and FLAG (Asp–Tyr–Lys–Asp– BU248722, BU397546, BU240892, CK607806, CK608165, Asp–Asp–Lys) tag sequence, and restriction sites for cloning, creating the BU241430, CV040649, BU299065, BU240964, CV889510, pcD2B-Ex2 vector. To facilitate cloning into a eukaryotic expression vector, pSELECT (InvivoGen), a third set of primers (D2B-pSEL-F1 and -R1) were BU235150, BU457064, BU215085, CF255243, BU473125, used to introduce BamHI and NheI sites for cloning. Two stop codons were CK610893, BU420073, and BU338155. The assembled pre- added at the 3′ end of the FLAG tag as well. The final expression vector was dicted DNase II sequence was used to design primers to amplify named pSELECT-D2B. 46 K.S. MacLea, H.H. Cheng / Gene 373 (2006) 44–51

KM-D2B-F1 (5′-GCGGCCGCACCATGACTGCGAGCTC Rockford, IL) and standardized by serial dilution of bovine TGTGTGGTGC-3′) serum albumen. KM-D2B-R1 (5′-CTCGAGCTTGTCGTCGTCGTCCTTGT AGTCTCCGGCTCCTATC CACGTGGAAGCATCATT-3′). 2.5. Western blot analysis D2B-pSEL-F1 (5′-TATAGGATCCACCATGACTGCGA GCTCTGTGTGGTGC-3′) Cell lysates were examined using the NuPAGE gel D2B-pSEL-R1 (5′-TATAGCTAGCTCAATCACTTGT electrophoresis system (Invitrogen). Lysates were boiled in CGTCGTCGTCCTTGTAGTCT CCGGC-3′). NuPAGE LDS buffer in the presence of β-mercaptoethanol (βME; Sigma) and electrophoresed on a 12% NuPAGE Bis– RNA was prepared using the RNAqueous kit (Ambion, Tris gel using NuPAGE 1× MOPS buffer. Proteins were Austin, Texas) with tissue from 11-day old ADOL Line 0 chick transferred to Immobilon-P polyvinylidene fluoride membrane embryos (Bacon et al., 2000). RACE-ready cDNA was (Millipore, Bedford, Massachusetts) and blocked in TBSTM prepared using the BD SMART RACE cDNA kit (BD [25mM Tris, pH 8, 125mM NaCl, 0.05% tween 20 (v/v), 5% Biosciences Clontech, Mountain View, California). cDNA non-fat dried milk] for 1h at room temperature. template was amplified using the Expand High FidelityPLUS The membrane was then probed with the anti-GFP rabbit PCR System (Roche Applied Science, Indianapolis, Indiana). polyclonal antibody (A-6455; Molecular Probes, Eugene, The eukaryotic expression vector into which the chicken DNase Oregon) at 1:3000, or the anti-FLAG mouse monoclonal II was cloned was pSELECT-zeo-mcs (InvivoGen, San Diego, antibody (M2; Sigma) at 1:500, in TBSTM for 3h, followed by California). For subcloning into pSELECT, digests of PCR goat anti-rabbit or -mouse secondary antibody conjugated to product and vector were undertaken with BamHI and NheI horseradish peroxidase (Zymax; Zymed, South San Francisco, enzymes (New England Biolabs (NEB), Beverly, Massachu- California) at a dilution of 1:3000 for 1h. Chemiluminescent setts) at 37°C for 1h in BamHI enzyme reaction buffer. The detection was performed with enhanced chemiluminescence fragments were ligated using the Rapid DNA Ligation Kit (ECL) reagent (Amersham, Piscataway, New Jersey). Protein (Roche) and the resulting vector was sequence verified and molecular weights were estimated by comparison to a standard named pSELECT-D2B. curve of sizes using the Precision Plus Protein Standards (Bio- As a control plasmid, the BamHI to XbaI (NEB) fragment of Rad, Hercules, California). pEGFP-N1 (BD Biosciences Clontech) containing enhanced green fluorescent protein (EGFP) coding sequence was ligated 2.6. Analysis of acid endonuclease activity in cell lysates into the BamHI and NheI sites of pSELECT-zeo-mcs. The control plasmid vector was designated pSELECT-EGFP. At DF-1 cells were lysed in AE5 buffer (see Section 2.4), snap each step of the process for both the experimental and control frozen, thawed and needle sheared 10 times through a 21-gauge vectors, sequence verification was undertaken using several needle. Lysates were assayed for acid endonuclease activity isolated clones on an ABI 3100 automatic DNA sequencer using the single radial enzyme diffusion (SRED) method (Applied Biosystems, Foster City, California). (Yasuda et al., 1992). In brief, of ethidium bromide- stained salmon sperm DNA in the pH 5 agarose gel assay plate 2.4. Cell culture and transient transfection was determined for each lysate sample by measurement of its diffusion diameter under UV illumination. Radial diffusion DF-1 chicken cells (Himly et al., 1998), an immortalized diameters were compared to serial dilution of bovine DNase II embryonic fibroblast cell line, were maintained in a mixture (Sigma). Lysate with equivalent total protein content was of Leibovitz L-15 and McCoy 5A media (1:1) supplemen- applied in 4μl to each well of the SRED assay plate and values ted with 2% fetal bovine serum (FBS), 1.1 g/L NaHCO3 are given in picogram bovine protein equivalent/nanogram total and penicillin–streptomycin (all from Sigma, St. Louis, protein. For each experiment, the SRED assay was performed in Missouri). DF-1 cells were cotransfected with either 4 μg five parallel wells and results are expressed as ±standard of pSELECT-D2B plasmid or no plasmid, together with deviation (SD) of the determinations. 1 μg of pSELECT-EGFP using Lipofectamine transfection reagent (Invitrogen, Carlsbad, California). After 36h, cells 3. Results and discussion were examined for transfectionefficiencybyfluorescence microscopy. For tunicamycin (TM) treatment, cells were 3.1. Identification of a DNase II homolog in expressed treated 24 h before harvest with 2.0 μg/ml TM (Sigma). All sequence tags cells were harvested using a cell scraper and pelleted. Pellets were washed once with cold PBS and resuspended Previous reports have identified an acid endonuclease in cold AE5 buffer: 0.1 M sodium acetate buffer containing activity in tissues from the chicken (G. gallus)(Torriglia et 20 mM EDTA, 1 mM phenylmethylsulfonylfluoride and al., 2001). Despite this, the only avian DNase that had been 1:100 protease inhibitor cocktail (P8340) (Sigma). Cells cloned and shown to have activity was a chicken homolog of the were lysed by snap-freezing in liquid nitrogen and were mammalian neutral-pH DNase I family named chicken needle-sheared using ten passes through a 21-gauge needle. deoxyribonuclease (Hu et al., 2003). However, our previous Protein content was determined using the BCA assay (Pierce, study of acid-pH DNase II homologs from various organisms K.S. MacLea, H.H. Cheng / Gene 373 (2006) 44–51 47 discovered a predicted partial sequence for chicken DNase II sequence. The chicken DNase II cDNA sequence was based on a few ESTs in the GenBank database (MacLea et al., subcloned into the pSELECT-zeo-mcs vector, creating the 2003a). The current study undertook BLAST searches of the pSELECT-D2B expression vector (Fig. 1). Its wild-type GenBank database, as well as the University of Manchester and translated protein sequence is shown in Fig. 2 along with University of Delaware chicken EST project databases, and partial and complete sequences from other avian and avian virus uncovered an even greater number of ESTs coding for an homologs (identified below). apparent homolog of DNase II in the chicken (see Section 2.3). The ESTs were isolated from a number of distinct tissues, 3.2. DNase II homologs in other avian and avian virus species including liver, ovary, fat, reproductive tract, brain, macro- phages and lymphoid tissue, Marek's disease virus-infected BLAST searches of the GenBank database utilizing either spleen, and chrondrocytes, as well as whole embryo and mixed human DNase IIα or the cloned chicken DNase II sequence as a adult tissues. When these fragments were assembled into a query sequence identified homologs from avian virus species, single sequence, the apparent full-length cDNA sequence of including the previously identified homolog in fowlpox virus chicken DNase II was present, as well as some of the 5′ and 3′ (CAA07012; Laidlaw et al., 1998) and an uncharacterized untranslated regions. homolog recently identified from canarypox virus (AAR83397; Custom oligonucleotide primers were designed to amplify Tulman et al., 2004). An additional BLAST search of the this region from freshly prepared cDNA. PCR amplification Songbird Neurogenomics Initiative EST project database with KM-D2B-F2 and-R2 primers yielded a fragment of the yielded a single EST hit with significant homology in the expected size, about 1100 bp. After the cloning of chicken zebra finch, Taeniopygia guttata (CK313786). DNase II and its subsequent sequence confirmation (100% We compared the observed protein sequence of chicken identity at the nucleotide level with the consensus sequence DNase II with the predicted fowlpox and canarypox sequences from assembled ESTs), PCR was used to introduce restriction and the partial zebra finch sequence (Fig. 2). The multiple sites and a DNA fragment coding for a FLAG tag (Asp–Tyr– sequence alignment demonstrates important regions in the Lys–Asp–Asp–Asp–Asp–Lys) at the 3′ end of the coding proteins, as observed in sequences cloned from other cloned

Fig. 2. Multiple sequence alignment of avian and avian virus DNase II homologs. Identity/similarity is represented with shading, with darker shading indicating greater degrees of similarity between residues. Conserved cysteine residues implicated in disulfide bridging are indicated with @ symbols above the residues in question, and additional cysteines conserved among the DNase IIβ subfamily of proteins are indicated by β. A putative catalytic residue, H302 in the chicken, which corresponds to the active H295 residue in the human DNase IIα sequence, is indicated with a star. In that portion of the alignment where sequence from all four species is available (S224–S360 of the chicken sequence), + symbols indicate positions where the bird residues or the virus residues, or both, are identical within their group (48/136 residues or 35.2%). − symbols indicate positions where identity is seen across groups; that is, where residues match between a bird species and a virus species (5/136 residues or 3.7%). 43/136 residues in this region (31.6%) are identical across all four species examined. 48 K.S. MacLea, H.H. Cheng / Gene 373 (2006) 44–51 species (MacLea et al., 2003a). In particular, the amino acid Fig. 2), comparison between the chicken and partial zebra finch H302 of the chicken sequence and the homologous residues in sequences and between the poxvirus sequences readily the other species appear to encode the expected catalytic demonstrates that the bird sequences and the virus sequences residue, by comparison with porcine and human DNases II are more similar within each group than between the groups. (Liao, 1985; MacLea et al., 2002). Further, recent computa- Indeed, in these 136 residues, we note 48 times (35.2%) where a tional protein fold prediction of the human protein (Cymerman residue at a given position in the bird or the virus sequences (or et al., 2005) would suggest that chicken DNase II residues both) are identical within the group but different from the other K304, D320, and N322 are also important in the catalytic group (annotated in Fig. 2 by a + character). However, in only function of the enzyme. Indeed, the presence of these residues in five cases (3.7%) did we note the opposite, where residues were each avian homolog provides additional evidence that these identical between a virus and a bird species but not within the residues may be significant. Similarly, the homologous cysteine bird group or the virus group (indicated by a − symbol in the residues predicted to be involved in disulfide bridging (MacLea figure). This quantitation lends credence to the hypothesis that et al., 2003b) are all present (excepting the missing piece of the the chicken and zebra finch DNases II are more similar to each zebra finch sequence) in these avian homologs. other than they are to either of the closely related poxvirus Noting differences between the avian and avian virus DNase DNase II sequences. II homologs may also be useful in analysis of the evolution of Because DNase II has been observed in most metazoan these family members in birds. Looking only at that portion of species, it is expected that the common ancestor of chicken and the alignment where we currently have sequence for all four zebra finch possessed one or more DNase II family members species (i.e., residues S224 to S360 of the chicken sequence; see that changed over time as species diverged. Since DNase II

Fig. 3. Analysis of DF-1 chicken cells transiently transfected with a plasmid encoding chicken DNase II. (A) Anti-FLAG and anti-GFP immunoblot of lysates from cells transfected with pSELECT-EGFP plasmid or pSELECT-EGFP cotransfected with pSELECT-D2B (expressing chicken DNase II-FLAG). The second and fourth lanes of the immunoblot show lysates from duplicate transfections in which cells were treated 24h before harvest with tunicamycin (TM) to inhibit N-glycosylation. Each lane represents equal total protein. (B) Acid endonuclease activity at pH 5 of the lysates from (A). Top, a representative image of the SRED assay gel for each lysate. Bottom, a graphical representation of the activity assay from experimental replicates. Values are given in picogram bovine protein equivalents/nanogram total protein. Error bars are ±SD, n=5. K.S. MacLea, H.H. Cheng / Gene 373 (2006) 44–51 49 homologs have only been identified in the two avipoxvirus enzyme. When acid endonuclease activity of protein from cells genomes (and in no other known virus), it can also be expected treated with TM was examined, it was observed that inhibition that a viral homolog was present in the common ancestor virus of N-glycosylation greatly diminished but did not completely of both fowlpox and canarypox (Tulman et al., 2004) after their ablate the activity of chicken DNase II, despite comparable divergence from the other chordopoxviruses. Indeed, when protein expression levels (Fig. 3B). Though some of the residual DNase II sequences from additional birds and avipoxvirus activity observed with TM treatment is probably due to species become available, sequence analysis may allow a very low level of glycosylated protein in lysates from treated determination of the likely common bird ancestor from which cells, the possibility of a slight intrinsic activity of protein the ancestral viral homolog was derived. produced in the absence of N-glycosylation cannot be ruled out. Overall, the expressed protein size, activity, and glycosyl- 3.3. Expression of active chicken DNase II ation profile of chicken DNase II are faithful to our current knowledge of DNase II structure and function. Despite Chicken DF-1 cells, which contain no endogenous acid significant sequence divergence between chicken DNase II endonuclease activity, were transiently transfected with pSE- and the human DNase IIα protein (37% identity at the amino LECT-D2B, the plasmid-expression construct encoding chicken acid level), it is clear that the enzymes carry out substantially DNase II tagged at its C-terminus with FLAG (Fig. 3). We similar enzymatic functions and are processed similarly at the concurrently examined the effect of the N-glycosylation inhibi- protein level. tor TM on both the size and activity of the expressed chicken DNase II. Cells that were transfected with pSELECT-D2B 3.4. DNase II gene structure and evolutionary history exhibited a prominent band at ∼45 kDa, which was reduced in size to ∼37 kDa after TM treatment (Fig. 3A), consistent with Though this research was begun before the recent comple- expected glycosylated and unglycosylated protein sizes, respec- tion of the first build of the chicken genome draft sequence tively (MacLea et al., 2002). The effect of inhibiting N-glyco- (Hillier et al., 2004), the conclusion of this important milestone sylation is also consistent with the observation that mammalian in chicken genomics means that it is now possible to easily DNase II is a glycoprotein (Liao, 1985; MacLea et al., 2002). examine the genomic structure of chicken DNase II. When the Another similarity of chicken DNase II with DNase II in chicken cDNA sequence is aligned against the draft sequence of mammals is the effect of TM treatment on activity of the chicken chromosome 8 (Fig. 4), several features of the gene

Fig. 4. Genomic structure of the chicken DNase II gene. The genomic structure is shown with a line representing non-coding sequence and solid boxes indicating coding regions. Unfilled boxes represent the incomplete 5′ and 3′ untranslated sequences cloned along with the cDNA. Numbers at the 5′ and 3′ ends of the gene are the chromosome 8 base positions of the first and last base included in the cloned sequence when superimposed on the first genome draft sequence of the chicken (Hillier et al., 2004). The exon and intron sizes are indicated in the left and right hand columns, respectively, for chicken DNase II and the DNase II human paralogs. 50 K.S. MacLea, H.H. Cheng / Gene 373 (2006) 44–51 structure are immediately apparent. First, the overall gene DNase II homologs (Table 1). DNase IIα is present on human structure seen in humans and mice (Krieser and Eastman, 1998; and its predicted counterpart in chicken should Krieser et al., 2001) is preserved: the gene contains six exons be found on chromosome 28. However, despite finding chicken and five introns. Though intron size varies substantially from homologs of genes that surround DNase IIα in the human or mammalian observations, the size of each exon is nearly rodent sequences, no complete or partial sequence of a chicken identical and all the splice sites seen in chicken are the same DNase IIα homolog can be found. This stands in contrast to ones observed in human and mouse DNase II genes. DNase IIβ homologs, which are readily identifiable in all of the In previous work on DNase II in human and rodent species, genomes, along with surrounding genes. the DNase IIα and DNase IIβ (or DLAD) subfamilies of this Given these results and the results of our sequence analysis, acid endonuclease were identified (Krieser and Eastman, 1998, it is likely that chicken DNase II is a homolog of the DNase IIβ Shiokawa and Tanuma, 1999). However, only one DNase II subfamily of DNase II acid endonucleases. However, because homolog had, at that time, been discovered in cow and pig, the no other DNase II-like EST or genomic sequences have yet been only other mammalian species in which DNase II had been found in chicken, this protein is simply named DNase II. studied (MacLea et al., 2003a). With recent elucidation of the Though it is possible that a DNase IIα homolog is present in chimpanzee (XP_512416, XP_524754), cow (XP_589701, chicken, the current lack of even a single EST sequence is XP_603463), and dog genomes (XP_533902, XP_537097), it suggestive of the absence of any other homologs. As the is clear that presence of two DNase II paralogs is a general genome draft sequence becomes more polished over time, it will characteristic of mammalian species. be possible to address this question definitively. The earlier evolutionary relationship of the DNase II family For the present, however, we can postulate that this gene is and timing of gene duplications is somewhat occluded, indeed the only homolog of DNase II found in the chicken and however, by the presence of three homologs in the nematodes that given this result, the DNase IIα and DNase IIβ subfamilies C. elegans and T. spiralis and the fish Takifugu rubpripes but appear to represent a gene duplication event from a single only one in the insects Drosophila and Anopheles, for example DNase II found in all amniotes, copied in the ancestors of the (MacLea et al., 2003a). This makes it difficult to make a mammals but not the sauropsids. Other gene duplications seen prediction as to the number of DNase II homologs that should in Takifugu and the nematode species, which do not easily fall be present in avian species. In the present study, we have into DNase IIα and DNase IIβ categories (MacLea et al., identified a gene from the chicken, G. gallus, which encodes the 2003a), may represent separate duplication events in these other first described avian DNase II. lineages. As shown in the analysis of the previously reported partial chicken sequence (MacLea et al., 2003a), the chicken DNase II 3.5. Conclusions protein sequence more closely resembles sequence of the DNase IIβ subfamily. The observation that chicken chromo- DNase II, a lysosomal and secreted endonuclease with an some 8 has conserved synteny with human acidic pH optimum, has been shown to be important in DNA across its entire length (Schmid et al., 2000; Hillier et al., 2004) fragmentation and degradation following cell death in normal further supports the claim that chicken DNase II is a DNase IIβ. mammalian development (Evans and Aguilera, 2003). Despite To examine whether there might be a DNase IIα homolog extensive biochemical characterization and its isolation from also present in the chicken, we undertook more careful analysis organisms as varied as nematodes and primates, no cloning, of regions within the chicken genome that are syntenic to expression, or activity of this enzyme has been reported in an regions of the human and rodent genomes known to contain avian species. In this paper, we have cloned and expressed

Table 1 Comparative arrangement of DNase IIα (D2A) and DNase IIβ(D2B) with nearby gene loci from syntenic regions of the human, mouse, rat, and chicken genomes Organism/chromosome Gene locia,b Human/19p13.2 PTPRS D2A KLF1 GCDH CALR Mouse/8 PTPRS D2A KLF1 GCDH CALR Rat/19q11 PTPRS D2A KLF1 GCDH CALR Chicken/28 Predictedc None Predicted None Predicted

Human/1p22.3 LPHN2 PRKACB UOXΨ D2B BXDC5 GNG5 SPATA1 CTBS Mouse/3 LPHN2 PRKACB UOX D2B BXDC5 GNG5 SPATA1 CTBS Rat/2q44 LPHN2 PRKACB UOX D2B BXDC5 GNG5 SPATA1 CTBS Chicken/8 Predicted Predicted Predicted D2 Predicted Predicted Predicted Predicted a PTPRS = protein tyrosine receptor type S; D2A = DNase II alpha; KLF1 = Kruppel-like factor 1; GCDH = glutaryl-CoA dehydrogenase; CALR = calreticulin. b LPHN2 = latrophilin 2; PRKACB = protein kinase, cAMP-dependent, catalytic, beta; UOX = uricase or urate oxidase; D2B = DNase II beta; BXDC5 = brix domain containing 5; GNG5 = guanine nucleotide binding protein gamma 5; SPATA1 = spermatogenesis associated 1; CTBS = di-N-acetyl-chitobiase. c Predicted = a homolog of the indicated gene has been predicted by computational analysis of the draft chicken genome sequence. However, unlike the corresponding genes in the human and rodent genomes, these genes have not yet been cloned. K.S. MacLea, H.H. Cheng / Gene 373 (2006) 44–51 51 chicken DNase II (Fig. 2), which demonstrates very similar Evans, C.J., Aguilera, R.J., 2003. DNase II: genes, enzymes and function. Gene – expression, structure, and function to that seen in mammalian 322, 1 15. Hillier, L.W., et al., 2004. Sequence and comparative analysis of the chicken systems (Fig. 3). Through comparison with other avian genome provide unique perspectives on vertebrate evolution. Nature 432, homologous sequences (Fig. 2) and with known gene structure 695–716. in other organisms (Fig. 4, Table 1), we have determined that Himly, M., Foster, D.N., Bottoli, I., Iacovoni, J.S., Vogt, P.K., 1998. The DF-1 chicken DNase II is properly designated a member of the DNase chicken fibroblast cell line: transformation induced by diverse oncogenes IIβ subfamily. Though it remains to be seen if a DNase IIα and cell death resulting from infection by avian leukosis viruses. Virology 248, 295–304. homolog will be identified in the chicken, initial data strongly Hu, C.C., Lu, S.C., Cheng, C.C., Chen, L.H., Liao, T.-H., 2003. Chicken suggests that chicken DNase II may be the only homolog deoxyribonuclease: purification, characterization, gene cloning and gene present in the chicken and therefore may represent the single- expression. J. Protein Chem. 22, 41–49. copy ancestral amniote form of the enzyme before the Kawane, K., et al., 2001. Requirement of DNase II for definitive erythropoiesis – evolutionary split between birds and mammals. in the mouse fetal liver. Science 292, 1546 1549. Krieser, R.J., Eastman, A., 1998. The cloning and expression of human Furthermore, the observation that mRNA molecules encod- deoxyribonuclease II: a possible role in apoptosis. J. Biol. Chem. 273, ing chicken DNase II were isolated from many different adult 30909–30914. and embryonic tissues (see Section 3.1), rather than from a Krieser, R.J., MacLea, K.S., Park, J.P., Eastman, A., 2001. The cloning, highly restricted set of tissues, may demonstrate an important genomic structure, localization, and expression of human deoxyribonuclease – evolutionary change between the avian and mammalian DNases II. Gene 269, 205 216. α Krieser, R.J., MacLea, K.S., Longnecker, D.S., Fields, J.L., Fiering, S., II. In mammals, DNase II is found in all tissues, while DNase Eastman, A., 2002. Deoxyribonuclease IIalpha is required during the IIβ is found chiefly in the developing lens and in a very phagocytic phase of apoptosis and its loss causes perinatal lethality. Cell restricted set of other tissues (Krieser et al., 2001; Nishimoto et Death Differ. 9, 956–962. al., 2003). If chicken DNase II represents the ancestral amniote Laidlaw, S.M., Anwar, M.A., Thomas, W., Green, P., Shaw, K., Skinner, M.A., DNase II form, these results indicate that the ancestral DNase II 1998. Fowlpox virus encodes nonessential homologs of cellular alpha- SNAP, PC-1, and an orphan human homolog of a secreted nematode protein. was ubiquitously expressed; however, gene duplication in the J. Virol. 72, 6742–6751. mammalian lineage eventually resulted in a partitioning of Liao, T.-H., 1985. The subunit structure and sequence of porcine expression pattern between the paralogs DNase IIα, which spleen deoxyribonuclease. J. Biol. Chem. 260, 10708–10713. retained universal expression, and DNase IIβ, which found its MacLea, K.S., Krieser, R.J., Eastman, A., 2002. Revised structure of the active expression greatly restricted. form of human deoxyribonuclease IIalpha. Biochem. Biophys. Res. Commun. 292, 415–421. MacLea, K.S., Krieser, R.J., Eastman, A., 2003a. A family history of Acknowledgements deoxyribonuclease II: surprises from Trichinella spiralis and Burkholderia pseudomallei. Gene 305, 1–12. We thank Laurie Molitor for excellent technical support and MacLea, K.S., Krieser, R.J., Eastman, A., 2003b. Structural requirements of Pam Campbell for isolation of ADOL Line 0 embryos. We are human DNase II alpha for formation of the active enzyme: the role of the signal peptide, N-glycosylation, and disulphide bridging. Biochem. J. 371, also grateful to Ronald Krieser of Bennington College and Alan 867–876. Eastman of Dartmouth College for valuable insights and Nishimoto, S., et al., 2003. Nuclear cataract caused by a lack of DNA discussions. This work was supported by funding from the degradation in the mouse eye lens. Nature 424, 1071–1074. United States Department of Agriculture Agricultural Research Schmid, M., et al., 2000. First report on chicken genes and chromosomes 2000. – Service. Cytogenet. Cell Genet. 90, 169 218. Shiokawa, D., Tanuma, S., 1999. DLAD, a novel mammalian divalent cation- independent endonuclease with homology to DNase II. Nucleic Acids Res. References 27, 4083–4089. Torriglia, A., Chaudun, E., Chany-Fournier, F., Courtois, Y., Counis, M.F., Bacon, L.D., Hunt, H.D., Cheng, H.H., 2000. A review of the development of 2001. Involvement of L-DNase II in nuclear degeneration during chick chicken lines to resolve genes determining resistance to diseases. Poult. Sci. retina development. Exp. Eye Res. 72, 443–453. 79, 1082–1093. Tulman, E.R., Afonso, C.L., Lu, Z., Zsak, L., Kutish, G.F., Rock, D.L., 2004. Baker, K.P., Baron, W.F., Henzel, W.J., Spencer, S.A., 1998. Molecular cloning The genome of canarypox virus. J. Virol. 78, 353–366. and characterization of human and murine DNase II. Gene 215, 281–289. Yasuda, T., Nadano, D., Sawazaki, K., Kishi, K., 1992. Genetic polymorphism Chenna, R., et al., 2003. Multiple sequence alignment with the Clustal series of of human deoxyribonuclease II (DNase II): low activity levels in urine and programs. Nucleic Acids Res. 31, 3497–3500. leukocytes are due to an autosomal recessive allele. Ann. Hum. Genet. 56, Cymerman, I.A., Meiss, G., Bujnicki, J.M., 2005. DNase II is a member of the 1–10. D superfamily. Bioinformatics 21, 3959–3962.