Molecular characterization and expression analysis of the gene coding for the porcine β3 integrin subunit (CD61)

Available online at www.sciencedirect.com

Gene 408 (2008) 9 – 17 www.elsevier.com/locate/gene

Molecular characterization and expression analysis of the gene coding for the porcine β3 integrin subunit (CD61) Angeles Jiménez-Marín a , Noemí Yubero a , Gloria Esteso a , Angela Moreno a , Juana Martín de las Mulas b , Luis Morera a , Diego Llanes a , Manuel Barbancho a , Juan J. Garrido a,⁎ a

Unidad de Marcadores Genéticos Moleculares, Departamento de Genética, Universidad de Córdoba, Campus de Rabanales, 14071 Córdoba, Spain b Departamento de Anatomía y Anatomía Patológica Comparadas, Facultad de Veterinaria, Universidad de Córdoba, Campus de Rabanales, 14071 Córdoba, Spain Received 11 May 2005; received in revised form 2 October 2007; accepted 10 October 2007 Available online 22 October 2007

Abstract Integrins are heterodimeric cell adhesion molecules with major roles in a variety of biological processes ranging from cell migration to tissue organization, immune and non-immune defense mechanisms and oncogenic transformation. Members of the β3 integrin subfamily are composed of a β3 subunit (CD61) non-covalently associated with two α subunits, αIIb (CD41) and αv (CD51), to constitute a group of transmembrane glycoproteins that participate in many physiologically important events. This investigation has focused on the molecular characterization of the cDNA encoding the porcine β3 integrin subunit. The deduced 762-amino acid sequence was 93, 92, 91, 89, 79 and 73% homologous to human, dog, rabbit, mouse, chicken and Xenopus laevis CD61 protein, respectively. Porcine CD61 molecule shares many structural features with human CD61, including a region containing a metal ion-dependent adhesion site (MIDAS) folding into an I domain-like structure. Through PCR-SSCP analysis and sequencing, six polymorphic positions were detected in the cDNA sequence of porcine CD61, and their frequencies were observed from a collection of 47 pigs. Expression analysis was done at two different levels: expression of the CD61 mRNA by RT-PCR and localization of the protein by immunohistochemistry. Our results show that CD61 transcripts were detected mainly in platelets and hematopoietic tissues. The immunohistochemical tissue localization of CD61 protein by a specific monoclonal antibody against CD61 recombinant protein showed that CD61 was expressed on vascular and non-vascular smooth muscle, epithelium and myeloid cells, being undetectable in cells of the lymphoid lineage. Furthermore, pulmonary intravascular macrophages (PIM), a subpopulation of macrophages which seem to play an important role in blood clearance, expressed much more CD61 when compared to pulmonary alveolar macrophages (PAM). The knowledge of the structure and distribution of the CD61 provides insight into the physiological function of the porcine β3 integrins and should be of importance in understanding the role of this integrin family in biological processes. © 2007 Elsevier B.V. All rights reserved. Keywords: Pig; Gene structure; Gene expression; SNP; RT-PCR; Immunohistochemistry

Abbreviations: DIG, digoxigenin; DMSO, dimethylsulfoxide; EDTA, ethylene diamine tetraacetic acid; FCS, fetal calf serum; HPA, human platelet alloantigen; mAb, monoclonal antibody; MIDAS, metal ion-dependent adhesion site; PBMC, peripheral blood mononuclear cells; PBS, phosphate buffered saline; PAM; pulmonary alveolar macrophages; PIM, pulmonary intravascular macrophages; PMSF, phenyl methyl sulphonyl fluoride; PRP, platelet-rich plasma; SNP, single nucleotide polymorphism; SSCP, single strand conformational polymorphism; RGD, arginine–glycine–aspartic acid; RT-PCR, reverse transcriptase polymerase chain reaction. ⁎ Corresponding author. Unidad de Marcadores Genéticos Moleculares, Departamento de Genética, Universidad de Córdoba, Campus de Rabanales, Edificio Gregor Mendel (C5), 14071 Córdoba, Spain. Tel.: +34 957 212692; fax: +34 957 218730. E-mail address: [email protected] (J.J. Garrido). 0378-1119/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2007.10.016

1. Introduction Integrins are a large family of heterodimeric transmembrane receptors functionally implicated in cell adhesion and recognition in a variety of biological processes so diverse as embryogenesis, hemostasis, tissue repair, immune response and metastatic diffusion of tumor cells (Hynes, 1987). They are cell-surface molecules formed by the association of one α and one β subunit in different combinations generating several receptor complexes with different expression patterns and

10

A. Jiménez-Marín et al. / Gene 408 (2008) 9–17

distinct ligand binding profiles (Hynes et al., 2002). Each integrin binds to only a limited series of ligands, ensuring that cell adhesion and migration are precisely regulated. Mammalian integrins have been divided into subfamilies according to their β subunit. Two members comprise the β3 subfamily of integrins, the platelet αIIbβ3 complex (CD41/ CD61, GP IIb/IIIa), and the more widely distributed vitronectin receptor αvβ3 complex (CD51/CD61) (Bosman, 1993). Integrin αIIbβ3 is the most abundant platelet adhesion receptor, which acts as a receptor for fibrinogen, fibronectin, and other RGDcontaining molecules. The primary function of this receptor is to bind fibrinogen on the surface of activated platelets. This activity results in platelet aggregation and in the formation of a platelet plug at the site of vessel injury (Shattil et al., 1985). Platelet aggregation is essential for normal hemostasis and depends on the capacity of αIIbβ3 to become activated and thus bind soluble fibrinogen with high affinity. Integrin αvβ3 is a receptor for a wide variety of extracellular matrix ligands with an exposed RGD sequence, including vitronectin, fibronectin, fibrinogen, thrombospondin, collagen, von Willebrand factor and osteopontin (Horton, 1997). This integrin is involved in signal transduction and cell to cell interactions and in the adult organism has been implicated in processes ranging from wound healing to tumor angiogenesis, arterial restenosis, osteoporosis and tumor progression (Brooks et al., 1994). In addition, different human viruses utilize αvβ3 integrin to mediate cellular internalization (Gavrilovskaya et al., 1998) and it was reported that foot-and-mouth disease virus, which is virulent for cattle and swine, can utilize the integrin αvβ3 as a receptor on cultured cells (Neff et al., 2000). Several mutations in the β3 chain of platelet αIIbβ3 cause Glanzmann thrombasthenia, an autosomal-recessive hemorrhagic disease characterized by defects in platelet function, such as the defective agonist-induced platelet aggregation and the absence or delay of clot retraction (French et al., 1998). Some of these genetic defects are large gene rearrangements or small deletions, but most are point mutations, which result in amino acid substitutions or abnormal mRNA splicing (Baker et al., 1997). Furthermore, CD61 is the most polymorphic integrin subunit in man, with at least seven alleles characterized at the molecular level. Some of the allelic isoforms are responsible for the immunopathologic syndromes neonatal alloimmune thrombocytopenia (NATP) and posttransfusion purpura (PTP), as well as for coronary artery disease (Bray et al., 1997). The domestic pig is generally accepted as an optimal experimental model in studies of cardiovascular disease because of its similarity to the human coronary circulation (Atar et al., 1994). The coronary anatomy of the swine mimics the human circulation closely, especially with regard to a relative absence of pre-existing collateral flow, which is crucial for studying myocardial infarction. In addition, swine and human platelet activation was similarly induced by a specific thrombin receptor. By contrast, platelets from other animal models like rabbits and rats lacked such a response (Kinlough-Rathbone et al., 1993). The purpose of the present study is to describe the molecular cloning and characterization of the porcine CD61 integrin gene and to compare the deduced amino acid sequence with its

homologous sequences previously reported for different species. Finally, RT-PCR was performed in parallel with immunohistochemistry to study the localization of porcine CD61 transcripts and the β3 integrins in a variety of porcine cells and tissues. 2. Materials and methods 2.1. Purification of porcine CD61 from platelets 5 mg of purified JM2E5, a monoclonal antibody specific for porcine αIIβ3 integrin (Pérez de la Lastra et al., 1997) were coupled to 1 ml of cyanogen bromide-activated Sepharose 4B beads (Pharmacia, Uppsala, Sweden) according to the manufacturer's indications. Lysis of the platelets was performed in 1% Nonidet P-40, 50 mM Tris, pH 8, 150 mM NaCl, 5 mM EDTA and 0.1 mM PMSF at 108 cells/ml for 1 h at 4 °C. Lysates were precleared by incubation for 24 h at 4 °C with normal mouse IgG-coupled Sepharose beads. Then, precleared lysates were incubated for an additional 24 h with JM2E5 mAb-coupled beads. After that, beads were washed several times with lysis buffer and, finally, with PBS. The adsorbed fraction was eluted from the beads by adding 50 mM ethylenamine, pH 11. The eluted fraction was dialyzed against PBS pH 7.2, concentrated and fractionated by SDS-PAGE on a 12% polyacrylamide gel under non-reducing conditions. A band with an apparent molecular weight of 105 kDa was excised and its N-terminal sequence was obtained by automated Edman degradation on an ABI 473A pulse-liquid phase protein sequencer (Applied Biosystems, Foster City, CA, USA). Peptide identity was analyzed on SwissProt, EMBL and GenBank databases. 2.2. RNA isolation from tissues and cells Porcine tissues were recovered from adult pigs immediately after slaughtering and frozen in liquid nitrogen until use. Porcine PBMC were isolated from heparinized whole blood by density gradient centrifugation on Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) at 900 g for 30 min. Mononuclear cells at the gradient interphase were collected by aspiration and washed twice in PBS. Porcine alveolar macrophages (PAM) were collected by bronchoalveolar lavage, washed with Hanks buffer containing 2 mM EDTA, resuspended at 5 × 107 cells/ml in FCS containing 10% DMSO and frozen in liquid nitrogen until use. Porcine platelet-rich plasma (PRP) was obtained by low centrifugation of blood anticoagulated with trisodium citrate in plastic tubes. Platelets were pelleted from PRP by centrifugation at 2200 g for 7 min and washed three times with PBS containing 5 mM EDTA. Total RNA from tissues and cells was purified according to the Tripure Isolation Reagent method (Roche, Basel, Switzerland) and quantified by absorption spectrometry at A260/A280. RNA samples were kept at − 80 °C after controlling the quality on a denaturing agarose gel. 2.3. Generation of a porcine-specific CD61 probe A CD61 cDNA probe was synthesized by reverse transcription and PCR amplification (RT-PCR) using total RNA from

A. Jiménez-Marín et al. / Gene 408 (2008) 9–17

spleen. 10 μg RNA, resuspended in 9.5 μl water, was heated for 3 min at 65 °C in the presence of 7.5 μM random hexamers (Pharmacia, Uppsala, Sweden), then cooled in ice. RNA was reverse transcribed using 200 units Moloney murine leukemia virus reverse transcriptase (Invitrogen, Barcelona, Spain) for 1 h at 42 °C in a final volume of 20 μl containing 4 μl of 5× reverse transcriptase buffer, 25 U ribonuclease inhibitor (Roche, Basel, Switzerland), 1 μl 20 mM dNTP and 2 μl 0.1 M dithiothreitol. At the end of the reaction, the solution was heated for 10 min at 95 °C and cooled in ice. 2 μl of this mixture was subjected to 35 cycles of PCR using Tth DNA polymerase (Biotools, Madrid, Spain) and two oligonucleotide primers: one deduced from the peptide sequence obtained by Edman degradation (5′-CCAACATCTGTGCCACACGAGG-3′, sense), and an additional oligonucleotide primer (5′-GGTCTTGGCATCAGTGG-3′, antisense) based on the published human CD61 cDNA sequence (Frachet et al., 1990). Each cycle consisted of incubations at 94 °C for 1 min, 55 °C for 1 min and 72 °C for 1 min. Finally the elongation was extended for 10 min at 72 °C. PCR band of the predicted size (based on the human CD61 sequence) was excised from low-melting agarose, purified using the Geneclean II purification system (Qbiogene, Carlsbad, CA, USA) and ligated into the pGEM-T vector (Promega, Madison, WI, USA) cloning site (insert:vector molar ratio 1:1). Transformation was done in competent XL1-blue Escherichia coli and the purified plasmids from several colonies were sequenced using ABI PRISM Terminator Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA). This strategy resulted in isolation of an internal 758 bp CD61 homologous. 2.4. Isolation of cDNA clones The 758 bp CD61 specific probe was labeled using PCR DIG Probe Synthesis Kit (Roche, Basel, Switzerland) according to the manufacturer's instructions. The screening of a porcine smooth muscle cDNA library (Stratagene, La Jolla, CA, USA) was performed as previously described (Sanz et al., 2007). Briefly, the transferred nylon membranes (Nytran N, Schleicher & Schuell, Dassel, Germany) were incubated overnight at 42 °C in hybridization solution (50% formamide, 5× SSC, 0.1% sarkosyl, 0.02% SDS, 1% blocking reagent, 10 ng/ml DIG-

11

labeled CD61 probe). After washing the membranes in 0.5× SSC, 0.1% SDS, twice, 15 min at 65 °C, the detection was performed using a colorimetric BCIP/NBT substrate method employing the DIG Detection Kit (Roche, Basel, Switzerland). Inserts from purified plaques were recovered as a pBluescript phagemid by in vivo excision from the parent lambda phage as described by Stratagene (La Jolla, CA, USA). 2.5. RT-PCR analysis RT-PCR was performed as described above, using 10 μg of RNA from lymphoid (spleen, thymus, PBMC, bone marrow, alveolar macrophages and platelets) and non-lymphoid (kidney and liver) tissues and gene-specific primers (5′-GAGAAGATTGGCTGGAGGAACGATGC-3′, sense and 5′-CTAAGTGCCCCGGTAGGTGATGTTGG-3′, antisense) to generate a cDNA band of 1590 bp. The amplifications were electrophoresed on 1% agarose/1× TAE gel, DNA blotted onto nylon membranes using a VacuGene XL (Pharmacia, Uppsala, Sweden) and fixed by baking for 2 h at 80 °C. The hybridization of the membrane with the purified 1590 bp PCR band used as homologous DIG-labeled probe and the detection were carried out as described for isolation the cDNA clones. Three replicates of RT-PCR analysis were carried out with no significant changes. 2.6. PCR-SSCP Six fragments of the porcine CD61 cDNA were amplified using genomic DNA samples from 47 unrelated Landrace × Large White pigs. DNA extraction from EDTA stabilized blood samples was carried out using the Eppendorf Perfect gDNA Blood Mini Kit (Eppendorf, Madrid, Spain). PCR primers and conditions for the amplification are given in Table 1. The obtained PCR products were electrophoretically analyzed on a GenePhor unit (Pharmacia, Uppsala, Sweden) and visualized by silver staining in order to identify the SSCP variants. Finally, PCR products containing probable mutations were purified using QIAquick PCR Purification Kit (Qiagen, Hilden, Germany) and sequenced in an ABI PRISM 3100 sequencer (Applied Biosystems, Foster City, CA, USA).

Table 1 Primers used for SNP identification Primer sequences 5′ → 3′

Location a

Size

F-GAGAAGATTGGCTGGAGGAAC R-ATGGTAGTGGAGGCAGAGTAATG F-GGCCTGGCTGGCTGGGGTTCC R-TTGCCCGTGATCTTGCCAAAGTCA F-CTTCTCAGCCTTTGTCCTCA R-CCCCCGCCGATTTCTCCT F-GCTCAGGCCTCTCGTTCC R-CCTGCCATCCTCACTCCCAAATAG F-TCAGGCCAAAGGAGAAATC R-AGAGGGTGGAATAGGAGGTAAAGA F-TCTTCTATTTGGGAGTGAGGATGG R-CTGGGACAAGGGGCAAAAG

Exon E

161

55

Exon I

169

62

3′-UTR

309

56

3′-UTR

281

57

3′-UTR

310

56

3′-UTR

321

57

a

(bp)

Annealing temperature (°C)

Location of the primers in exonic sequences was determined by comparison with the corresponding regions in human CD61 gene (Lanza et al., 1990).

12

A. Jiménez-Marín et al. / Gene 408 (2008) 9–17

2.7. Recombinant CD61 protein (rpCD61) DNA encoding an extracellular domain of the porcine CD61 corresponding to the putative integrin ligand binding region was amplified by PCR from plasmid template using the following primers: 5′-GGTCGGATCCCAACATCTGTGCCACGCGAGG-3′ (sense) and 5′-TCTAAGCTTGGTCTTGGCATCAGTGG-3′ (antisense). These primers contained restriction sites (indicated in bold) enabling ligation into the expression vector pET28b (Novagen, Darmstadt, Germany) following digestion of the PCR product and the vector with BamHI and HindIII. The resulting recombinant plasmids were used to transform Escherichia coli strain BL21 (DE3) (Novagen, Darmstadt, Germany). Protein expression was induced by the addition of 1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) for 4 h at 37 °C. Cells were collected by centrifugation at 4000 g and suspended in a buffer containing 0.1 M Na-phosphate, 0.01 M Tris/HCl (pH 8) and 8 M urea, then sonicated three times at intervals of 30 s on ice. A 27 kDa recombinant His-Tag protein was purified from the insoluble inclusion body by affinity chromatography using the one step QIAexpress NiNTA agarose matrix (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Purified recombinant protein was dialyzed against PBS containing 5 M urea and adjusted to a final concentration of 1 mg/ml. Purity of the recombinant protein was assessed by SDS-PAGE, followed by Coomassie blue staining. Protein identification was carried out by MALDITOF mass spectrometry analysis of the tryptic peptides.

The sections were deparaffinized and rehydrated in xylene and descending concentrations of ethanol, respectively. Endogenous peroxidase activity was inhibited by treatment with 3% hydrogen peroxidase in distilled water for 30 min at room temperature. After washing with PBS, the sections were incubated with normal goat serum (1:10 dilution in PBS) (Vector, Burlingame, CA, USA) for 30 min at room temperature. After removing the serum, mAb AJ2G11 supernatant (1/1000 dilution in PBS) or an irrelevant mAb (as negative control) were added for 18 h at 4 °C in a wet chamber. The sections were washed in PBS and incubated with biotinylated anti-mouse Ig (Dako, Glostrup, Denmark) diluted 1/ 50 in PBS for 30 min at room temperature. After washing again in PBS, tissue sections were covered with avidin–biotin–peroxidase complex (Sigma, Barcelona, Spain) diluted 1/50 with PBS for 1 h in a wet chamber at room temperature, washed and then developed with 3, 3′-diaminobenzidine (Sigma, Barcelona, Spain) (5 μg in 10 ml PBS). Sections were counterstained with Mayer hematoxylin and mounted with Eukitt. The sections of lymph node were covered with biotin–streptavidin–alkaline phosphatase complex (Biogenex, San Ramon, CA, USA). The chromogen substrate for alkaline phosphatase was naphthol and Fast Red (20% w/v). After incubation, slides were washed and mounted in aqueous medium of Shandon Immuno-Mount (Sigma, Barcelona, Spain). 3. Results and discussion 3.1. Purification of porcine CD61 and N-terminal protein sequence

2.8. Monoclonal antibody production AJ2G11 monoclonal antibody was produced using previously described immunization and cells fusion procedures (Arce et al., 2002). Briefly, female BALB/c mice were immunized with 50 μg of rpCD61. Spleen cells from immune mice were fused with Sp2/0 myeloma cells. Hybridoma clones were selected on the basis of binding secreted antibody to rpCD61 by indirect ELISA. Antibody-producing hybridomas reacting positively were cloned at least twice by limiting dilution. Immunoglobulin classes and subclasses were determined in solid-phase ELISA using rabbit antisera specific for mouse heavy and light chains and a peroxidase-conjugated goat antirabbit immunoglobulin (Sigma, Barcelona, Spain). Monoclonal antibody AJ2G11 was of the IgM subclass.

Platelets from porcine PRP were used as a source for the purification of the JM2E5 antigen. By passing a platelet lysate over a JM2E5 mAb immunoaffinity column we isolated a protein that fractionated under no reducing conditions by 12% SDS-PAGE showing two bands of 140 kDa and 105 kDa, corresponding to the approximate molecular weight of the integrin α and β chains, respectively. After purification, the isolated 105 kDa protein was analyzed on an automate sequencer and its resultant N-terminal sequence was Gly-ProAsn-Ile-X-Ala-Thr-Arg-Gly-Val. Sequence comparison with Nterminus from human CD61 protein (Frachet et al., 1990) showed a 80% of amino acid identity confirming that the isolated 105 kDa precipitated protein by JM2E5 monoclonal antibody corresponded to the porcine CD61 molecule.

2.9. Immunohistochemistry

3.2. Porcine CD61 sequence

Spleen, thymus, bone marrow, lymph node and lung from healthy animals and spleen from a pig with Glasser disease were studied. All tissue specimens were fixed in Bouin liquid for 16 h. Tissues were dehydrated in ascending concentrations of ethanol and xylene and embedded in paraffin. Sections of 5 μm were prepared from selected tissue blocks and consecutive sections were placed on slides coated with Vectabound (Vector, Burlingame, CA, USA). The tissue slides were kept at 55 °C for 45 min in an oven to improve the adherence of sections to glass.

The cDNA sequence encoding the porcine CD61 was isolated by screening a porcine smooth muscle lambda cDNA library under stringent conditions with a 758-bp specific probe. The nucleotide and deduced amino acid sequences of the porcine CD61 cDNA have been deposited at GenBank under accession number AF282890. Sequence analysis revealed a 4566-bp cDNA containing a 31-bp untranslated 5′ flanking sequence, a single open reading frame of 2352 bp encoding a polypeptide of 784 amino acids and a 2183-bp untranslated 3′ flanking region. The first 22 amino acids of the translate protein

A. Jiménez-Marín et al. / Gene 408 (2008) 9–17

13

Fig. 1. Alignment of the porcine CD61 protein with other orthologous CD61 protein sequences. The sequences were derived from GenBank entries with accession numbers AAA35927 (human), NP_001003162 (dog), AAD51955 (rabbit), NP_058060 (mouse), CAA51069 (chicken) and AAA17427 (Xenopus laevis). The signal peptide sequences are not included in the comparison. The stripes above the sequences represent the deduced different constitutive parts of the protein:extracellular domain ( ), transmembrane region ( ), cytoplasmic tail ( ), I-like domain ( ) cysteine-rich repeats ( ), and MIDAS-like motif and fibrinogen binding site ( ). Potential N-glycosylation sites and cysteine residues are marked in dark grey and light grey, respectively. Essential amino acid residues in MIDAS-like stretch and fibrinogen binding site domain have a black background. Potential cytoplasmic-tail phosphorylation sites are marked with an asterisk at the bottom of the sequences.

14

A. Jiménez-Marín et al. / Gene 408 (2008) 9–17

Table 2 Percent sequence identity between porcine (Po) CD61 constitutive blocks with those from human (Hu), dog (Ca), rabbit (Ra), mouse (Mo), chicken (Ch) and Xenopus laevis (Xe) Block

Po vs. Hu (%)

Po vs. Ca (%)

Po vs. Ra (%)

Po vs. Mo (%)

Po vs. Ch (%)

Po vs. Xe (%)

Overall Extracellular domain I-like domain MIDAS-like motif Transmembrane region Cytoplasmic region

93 92 95 100 97 100

92 92 95 100 100 100

91 92 95 100 93 100

89 89 93 100 100 100

79 80 88 100 97 88

73 73 77 100 90 88

are predominantly hydrophobic and correspond to the signal peptide sequence. Excluding this sequence, the porcine mature CD61 consists of 762 amino acids with a predicted molecular weight of 86.3 kDa and 8 potential N-glycosylation sites (AsnX-Ser/Thr). If carbohydrate chains with an average molecular weight of 2.5 kDa are assumed to attach to all putative glycosylation sites, the total molecular weight of the mature CD61 molecule would be 106.3 kDa. This value is consistent with the estimated size of the pig CD61 on SDS-PAGE. Translate mature porcine sequence conserves the main structural characteristics that define the function of CD61 in other species (Fig. 1). Thus, the extracellular domain contains an inserted I-like domain of 260 amino acids (residues 111–370) homologous to the A domain of von Willebrand factor. This I domain is also found in a subset of α integrin subunits, adopts a classical α,β Rossmann fold and includes a putative metal iondependent adhesion site (MIDAS) DXSXS motif (residues 119– 123) that is critical for the RGD-ligand binding function of the receptor (Tozer et al., 1996). Located in the carboxy-terminal portion there is a stretch of 29 hydrophobic amino acids that likely constitutes the transmembrane domain. Following the putative transmembrane domain there is a 41-amino acid sequence that probably represents the cytoplasmic domain of the molecule. This cytoplasmic segment contains two phosphorylation sites at positions 747 (tyrosine) and 753 (threonine) also conserved in other β-integrin subunits (Pasqualini and Hemler, 1994). The deduced amino acid sequence of the processed swine CD61 has high cysteine content (7.1%), having a total of 56 cysteine residues. Similar to the porcine β1 integrin subunit (Jiménez-Marín et al., 2000), many of these cysteines (32 residues) are arranged in four cysteine-rich, tandemly repeated domains of about 40 residues each and located in an amino acid segment adjacent to the transmembrane domain. 3.3. Comparison of porcine CD61 amino acid sequence with other CD61 molecules The deduced protein sequence of the porcine CD61 was compared to the sequences of six different species. The porcine CD61 protein is highly conserved, having overall 93%, 92%, 91%, 89%, 79%, and 73% amino acid identity with human, dog, rabbit, mouse, chicken and Xenopus proteins, respectively, with the highest identity for the MIDAS-like motif (100%) and the lowest for the extracellular domain (86% average) (Table 2). As shown in Fig. 1, the porcine CD61 protein shares common structural and functional elements with CD61 molecules from the

other species. Also conserved among the CD61 sequences are the number and identical positions of the cysteine residues, which is consistent with a role in maintaining the global structure of the protein. By comparison with the human sequence, two potential sites of phosphorylation are conserved in the porcine cytoplasmic region, residues Tyr747 and Thr753, which is according with their role in the mechanism of modulation of the avidity of the integrins for their ligands and in the interactions of these membrane receptors with the cellular cytoskeleton (Lerea et al., 1999). The very high interspecies conservation of the putative MIDAS-like and I-like domain confirms their essential role in the β3 integrin's function. In effect, mutational studies have shown that the metal-coordinating consensus sequence DXSXS has a critical functional role in the RGD-ligand binding function of αIIbβ3 (Bajt and Loftus, 1994). In this context, the high degree of conservation of these residues in the porcine sequence suggest that, as in the human case, the RGD-ligands could interact with divalent cations bound to this site in the β3 subunit. A high level of conservation is observed in a second potential ligand-interactive site of β3 integrin, defined in human by residues Ser211-Gly222. Peptides corresponding to this sequence bound fibrinogen and blocked its binding to αIIbβ3 (Charo et al., 1991). Inside this sequence, Asp217 and Glu220 were reported as residues essential for ligand binding function of αIIbβ3 (Tozer et al., 1996). Fig. 1 shows that these residues are conserved in all the CD61 amino acid sequences analyzed. In humans, polymorphisms at amino acid positions 33, 40, 143, 407, 489 and 636 of the mature β3 integrin subunit are immunogenic and have been associated with increased risk of myocardial infarction and coronary stent thrombosis (Weiss et al., 1996; Walter et al., 1997). The deduced amino acid sequence of porcine β3 integrin corresponds to the highly frequent allelic form for five of the six alloantigenic sites of human platelets: Leu33, Leu40, Arg143, Pro407, and Arg636. Interestingly, the amino acid in position 489 (HPA-6 alloantigen) of porcine β3 subunit, glutamine, differs from the amino acid arginine, present in highest frequency in humans. This HPA-6 platelet antigen corresponds to the Arg489 (HPA6a) → Gln489 (HPA-6b) polymorphism and has been associated with NATP in humans (Nurden, 1995). 3.4. SNP identification In order to search for SNPs in the porcine CD61 gene, genomic DNA sequences from 47 unrelated Landrace × Large White pigs were investigated. Primers used to amplify six

A. Jiménez-Marín et al. / Gene 408 (2008) 9–17

genomic fragments are in Table 1. The PCR products were used for SSCP analysis to detect the existence of DNA variation and the sequence polymorphism was confirmed through DNA sequencing. Six SNPs were identified and their nucleotide sequence locations and allele frequencies are summarized in Table 3. Two SNPs are located in the open reading frame (ORF), precisely within the exons E (position 847) and I (position 1609) by correspondence with human genomic sequence (Lanza et al., 1990). These SNPs are conservative mutations causing no change in the amino acid sequence. The other four SNPs are located within the 3′ untranslated region of the gene. 3.5. Expression of CD61 in porcine tissues and cells To investigate the pattern of porcine CD61 mRNA expression, RT-PCR analysis was conducted with a variety of adult pig tissues using gene-specific primers to generate a cDNA band of 1590 bp. To confirm the specificity of the amplification, a Southern blot of the PCR products was probed with the purified PCR 1590 bp band which was expected to be specific to CD61. As shown in Fig. 2, CD61 expression was observed in all lymphoid tissues and cells analyzed. The highest level of CD61 expression was observed in platelets and in PBMC and the lowest expression was found in bone marrow cells. However, CD61 mRNA was not detected in liver cells. In kidney, a slightly darker smear is visualized at the CD61 position indicating that it is likely that the gene was lightly expressed in this tissue according with our previous findings observed in porcine tissues (Moreno et al., 2002). The precise localization of the protein coded by CD61 was studied by immunohistochemistry with AJ2G11 monoclonal antibody developed using a CD61 recombinant protein. The antibody was raised against a recombinant protein corresponding to a 243-amino acid polypeptide fragment located in the putative ligand binding domain of the porcine CD61 molecule. The reactivity of this monoclonal antibody was tested on formalin-fixed, paraffin-embedded sections of the following porcine tissues: spleen, thymus, bone marrow, lymph node and lung. Representative results of this immunohistochemical analyses are shown in Fig. 3. In the spleen, the monoclonal antibody reacted only with splenic macrophages (Fig. 3a). Lymphoid cells in white pulp were negative. In the thymus, the positive reaction was mainly found in the epithelial reticular

Table 3 Single nucleotide polymorphisms within the porcine CD61 gene Nucleotide Substitution

Position a

Allele frequencies

T→C C→T A→T C→G A→G G→C

847 1609 2728 2925 2984 3089

0.79/0.21 0.73/0.27 0.81/0.19 0.64/0.36 0.62/0.38 0.69/0.31

a Numbering is according to the deposited sequence (GenBank accession no. AF282890).

15

Fig. 2. Expression patterns of CD61 transcripts in different porcine cells and tissues. Thymus (a), alveolar macrophages (b), liver (c), spleen (d), bone marrow (e), platelets (f), kidney (g), and PBMC (h). 18S amplification was used as control.

cells forming the medullary Hassall's corpuscles (Fig. 3b). Cortical thymocytes were consistently negative. In the bone marrow, the integrin expression was mostly restricted to the megakaryocytes and hematopoietic cells of myeloid lineage (Fig. 3c). In lymph node, the monocytes and polymorphonuclear leukocytes were clearly labeled, whereas lymphoid cells were negative (Fig. 3d). The smooth muscle layers in the blood vessel walls are also intensely stained. In lung (Fig. 3e) the reaction was positive in circulating monocytes and in fusiform cells apposed to the capillary endothelium which were identified as pulmonary intravascular macrophages (PIM). However, CD61 was undetectable in pulmonary alveolar macrophages (PAM). Finally the specificity of AJ2G11 was confirmed using samples of spleen from a pig with Glasser disease, a generalized Haemophilus parasuis infection causing important vascular lesions and unusual presence of megakaryocytes in spleen. Fig. 3f shows that the expression of the β3 integrin in this tissue was restricted to the connective tissue of the trabecula, isolated macrophages and mature megakaryocytes liberating platelets in the splenic red pulp. As in other tissues, lymphoid cells were negative. Taken together, our results indicate that, in pig, the histological distribution of the integrin β3 subunit is similar to that of its homologous human molecule. Thus, the expression of the CD61 protein in porcine vascular and non-vascular smooth muscle could be due to the presence of αvβ3 as it was reported in previous studies in human (Yee et al., 1998). Similarly, the positive reaction found in myeloid cells would be expected on the basis of the presence of αIIbβ3 in human platelets and megakaryocytes (Heijnen et al., 1998) and αvβ3 in monocytes, macrophages and granulocytes from different species (Singh et al., 2001). The fact that the expression of CD61 was not detectable on PAM, although it was detected by RT-PCR, can be explained by the probable low level of expression of pig αvβ3 integrin on the surface of normal PAM, as has been described in human lung (Conron et al., 2001; Pons et al., 2005). On the other hand, in humans, upregulation of αvβ3 has been associated to enhanced susceptibility of PAM to adenovirus infection (Conron et al., 2001). Therefore, our observation that PAM express less αvβ3 compared to both monocytes, the circulating precursors of PAM (Andreesen et al., 1990), and PIM, a highly immunoreactive macrophage population apposed to lung capillary endothelium (Chitko-McKown and Blecha, 1992), can have important significance to understand the roles

16

A. Jiménez-Marín et al. / Gene 408 (2008) 9–17

Fig. 3. Immunohistochemical staining of formalin-fixed, paraffin-embedded sections of different porcine tissues with AJ2G11 monoclonal antibody. (a) Spleen (40×). Positive reaction in splenic macrophages. (b) Thymus (40×). Epithelial cells forming the outer layer of the Hassall's corpuscles (HC) were positive. Cortical (C) and medullary (M) thymocytes are negative. (c) Bone marrow (40×). Positive staining of megakaryocytes (MK) and hemopoietic cells of myeloid lineage. (d) Lymph node (40×). Monocytes, polymorphonuclear leukocytes and vascular smooth muscle were stained with CD61 monoclonal antibody. Mononuclear lymphocytes were negative. (e) Lung (40×). Circulating monocytes (CM) and pulmonary intravascular macrophages (PIM) are positive. Pulmonary alveolar macrophages (PAM) were negative. (f) Spleen from a pig with Glasser disease (40×). Mature megakaryocytes (MK) and connective tissue of the trabecula (T) were positive and lymphoid cells were negative.

of αvβ3, including phagocytic and cytotoxic activities (ChitkoMcKown et al., 1991; Pons et al., 2005). Finally, our immunohistochemical study shows that there was no detectable expression of β3 integrin receptor on lymphoid cells in the spleen, thymus, bone marrow and lymph node. These results are in agreement with previous findings observed in human tissues (Zutter, 1991).

3.6. Conclusions In conclusion, here we report the isolation and sequencing of the porcine β3 integrin subunit (CD61) cDNA, demonstrating that is a highly conserved molecule that shares the main characteristics of the β3 integrins from other species. In addition, we have studied the expression of the β3 integrin

A. Jiménez-Marín et al. / Gene 408 (2008) 9–17

subunit with the aim of defining its detailed topographic distribution in porcine tissues and cells. Our results are of particular interest because the pig is an animal model system for a variety of immunological and pathological studies. Thus, β3 integrins have a differential expression in normal tissues and tumors and several authors have found that the altered expression of integrins on tumor cells can change their adhesive properties and biological behavior (Varner and Cheresh, 1996). On the other hand, our work contributes to the knowledge about the similarity between swine and human platelet surface receptors, key information in the context of the pig as a model to study coronary artery disease. Acknowledgments The authors thank Reyes Álvarez and María Friend for excellent technical assistance. This work was supported by the National R&D Program Grant of the Spanish Ministry of Education and Science (AGL2002-00529 and AGL200501561). AJM is a postdoctoral researcher supported by an EADGENE fellowship. NY and GE are predoctoral research fellows of the University of Córdoba. JJG was a recipient of a “Ramón y Cajal” grant of the Spanish Ministry of Education and Science. References Andreesen, R., et al., 1990. Surface phenotype analysis of human monocyte to macrophage maturation. J. Leukoc. Biol. 47, 490–497. Arce, C., Moreno, A., Millán, Y., Martín de las Mulas, J., Llanes, D., 2002. Production and characterization of monoclonal antibodies against dog immunoglobulin isotypes. Vet. Immunol. Immunopathol. 88 (1–2), 31–41. Atar, D., Serebruany, V., Poulton, J., Godard, J., Schneider, A., Herzog, W.R., 1994. Effects of magnesium supplementation in a porcine model of myocardial ischemia and reperfusion. J. Cardiovasc. Pharmacol. 24, 603–611. Bajt, M.L., Loftus, J.C., 1994. Mutation of a ligand binding domain of β3 integrin. Integral role of oxygenated residues in αIIbβ3 (GPIIb–IIIa) receptor functions. J. Biol. Chem. 269, 20913–20919. Baker, E.K., Tozer, E.C., Pfaff, M., Shattil, S.J., Loftus, J.C., Ginsberg, M.H., 1997. A genetic analysis of integrin function: Glanzmann thrombasthenia in vitro. Proc. Natl. Acad. Sci. U. S. A. 94 (5), 1973–1978. Bosman, F.T., 1993. Integrins: cell adhesives and modulators of cell function. Histochem. J. 25, 469–477. Bray, P.F., Weiss, E.J., Tayback, M., Goldschmidt-Clermont, P.J., 1997. PlA1/ A2 polymorphism of platelet glycoprotein IIIa and risk of cardiovascular disease. Lancet 349 (9058), 1100–1101. Brooks, P.C., et al., 1994. Integrin αvβ3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 79, 1157–1164. Charo, I.F., Nannizzi, L., Phillips, D.R., Hsu, M.A., Scarborough, R.M., 1991. Inhibition of fibrinogen binding to GP IIb–IIIa by a GP IIIa peptide. J. Biol. Chem. 266, 1415–1421. Chitko-McKown, C.G., Blecha, F., 1992. Pulmonary intravascular macrophages: a review of immune properties and functions. Ann. Rech. Vet. 23, 201–214. Chitko-McKown, C.G., Chapes, S.K., Brown, R.E., Philips, R.M., McKown, R.D., Blecha, F., 1991. Porcine alveolar and pulmonary intravascular macrophages: comparison of immune functions. J. Leukoc. Biol. 50, 364–372. Conron, M., et al., 2001. Alveolar macrophages and T cells from sarcoid, but not normal lung, are permissive to adenovirus infection and allow analysis of NF-kB-dependent signalling pathways. Am. J. Respir. Cell Mol. Biol. 25, 141–149. Frachet, P., Uzan, G., Thevenon, D., Denarier, E., Prandini, M.H., Marguerie, G., 1990. GPIIb and GPIIIa amino acid sequences deduced from human megakaryocyte cDNAs. Mol. Biol. Rep. 14 (1), 27–33.

17

French, D.L., et al., 1998. Prenatal diagnosis of Glanzmann thrombasthenia using the polymorphic markers BRCA1 and THRA1 on chromosome 17. Br. J. Haematol. 102 (2), 582–587. Gavrilovskaya, I.N., Shepley, M., Shaw, R., Ginsberg, M.H., Mackow, E.R., 1998. β3 integrins mediate the cellular entry of hantaviruses that cause respiratory failure. Proc. Natl. Acad. Sci. U. S. A. 95 (12), 7074–7079. Heijnen, H.F., Debili, N., Vainchencker, W., Breton-Gorius, J., Geuze, H.J., Sixma, J.J., 1998. Mutivesicular bodies are an intermediate stage in the formation of platelet α-granules. Blood 91, 2313–2325. Horton, M.A., 1997. The αvβ3 integrin “vitronectin receptor”. Int. J. Biochem. Cell Biol. 29, 721–725. Hynes, R.O., 1987. Integrins: a family of cell surface receptors. Cell 48, 549–554. Hynes, R.O., et al., 2002. The diverse roles of integrins and their ligands in angiogenesis. Cold Spring Harbor Symp. Quant. Biol. 67, 143–153. Jiménez-Marín, A., Garrido, J.J., de Andrés-Cara, D.F., Morera, L., Barbancho, M.J., Llanes, D., 2000. Molecular cloning and characterization of the pig homologue to human CD29, the integrin β1 subunit. Transplantation 70 (4), 649–655. Kinlough-Rathbone, R.L., Rand, M.L., Packham, M.A., 1993. Rabbit and rat platelets do not respond to thrombin receptor peptides that activate human platelets. Blood 82, 103–106. Lanza, F., Kieffer, N., Phillips, D.R., Fitzgerald, A., 1990. Characterization of the human platelet glycoprotein IIIa gene. Comparison with the fibronectin receptor β-subunit gene. J. Biol. Chem. 265, 18098–18103. Lerea, K.M., Cordero, K.P., Sakariassen, K.S., Kirk, R.I., Fried, V.A., 1999. Phosphorylation sites in the integrin β3 cytoplasmic domain in intact platelets. J. Biol. Chem. 22, 1914–1919. Moreno, A., Lucena, C., López, A., Garrido, J.J., Pérez de la Lastra, J.M., Llanes, D., 2002. Immunohistochemical analysis of beta3 integrin (CD61): expression in pig tissues and human tumors. Histol. Histopathol. 17 (2), 347–352. Neff, S., Mason, P.W., Baxt, B., 2000. High-efficiency utilization of the bovine integrin αvβ3 as a receptor for foot-and-mouth disease virus is dependent on the bovine β3 subunit. J. Virol. 74 (16), 7298–7306. Nurden, A.T., 1995. Polymorphisms of human platelet membrane glycoproteins: structure and clinical significance. Thromb. Haemost. 74, 345–351. Pasqualini, R., Hemler, M.E., 1994. Contrasting roles for integrin β1 and β5 cytoplasmic domains in subcellular localization, cell proliferation, and cell migration. J. Cell Biol. 125, 447–460. Pérez de la Lastra, J.M., Moreno, A., Pérez, J., Llanes, D., 1997. Characterization of the porcine homologue to human platelet glycoprotein IIb–IIIa (CD41/ CD61) by a monoclonal antibody. Tissue Antigens 49 (6), 588–594. Pons, A.R., Noguera, A., Blanquer, D., Sauleda, J., Pons, J., Agustí, A.G.N., 2005. Phenotypic characterisation of alveolar macrophages and peripheral blood monocytes in COPD. Eur. Respir. J. 25, 647–652. Sanz, G., et al., 2007. Molecular cloning, chromosomal location and expression analysis of porcine CD14. Dev. Comp. Immunol. 31, 738–747. Shattil, S.J., Brass, L.F., Bennett, J.S., Pandhi, P., 1985. Biochemical and functional consequences of dissociation of the platelet membrane glycoprotein IIb–IIIa complex. Blood 66 (1), 92–98. Singh, B., Rawlings, N., Kaur, A., 2001. Expression of integrin αvβ3 in pig, dog and cattle. Histol. Histopathol. 16, 1037–1046. Tozer, E.C., Liddington, R.C., Sutcliffe, M.J., Smeeton, A.H., Loftus, J.C., 1996. Ligand binding to integrin αIIbβ3 is dependent on a MIDAS-like domain in the β3 subunit. J. Biol. Chem. 271 (36), 21978–21984. Varner, J.A., Cheresh, D.A., 1996. Integrins and cancer. Curr. Opin. Cell Biol. 8 (5), 724–730. Walter, D.H., Schächinger, V., Elsner, M., Dimmeler, S., Zeiher, A.M., 1997. Platelet glycoprotein IIIa polymorphisms and risk of coronary stent thrombosis. Lancet 350, 1217–1219. Weiss, E.J., et al., 1996. A polymorphism of a platelet glycoprotein receptor as an inherited risk factor for coronary thrombosis. N. Engl. J. Med. 334, 1090–1094. Yee, K.O., Rooney, M.M., Giachelli, C.M., Lord, S.T., Schwartz, S.M., 1998. Role of the β1 and β3 integrins in human smooth muscle cell adhesion to and contraction of fibrin clots in vitro. Circ. Res. 83, 241–251. Zutter, M.M., 1991. Immunolocalization of integrin receptors in normal lymphoid tissues. Blood 77, 2231–2236.