Xenotransplantation 2014: 21: 543–554 doi: 10.1111/xen.12124

© 2014 The Authors. Xenotransplantation Published by John Wiley & Sons Ltd XENOTRANSPLANTATION

Original Article

Cloning and expression of porcine b1,4 N-acetylgalactosaminyl transferase encoding a new xenoreactive antigen Byrne GW, Du Z, Stalboerger P, Kogelberg H, McGregor CGA. Cloning and expression of porcine b1,4 N-acetylgalactosaminyl transferase encoding a new xenoreactive antigen. Xenotransplantation 2014; 21: 543–554. © 2014 The Authors. Xenotransplantation Published by John Wiley & Sons Ltd Abstract: Background: Xenograft rejection of pigs organs with an engineered mutation in the GGTA-1 gene (GTKO) remains a predominantly antibody mediated process which is directed to a variety of nonGal protein and carbohydrate antigens. We previously used an expression library screening strategy to identify six porcine endothelial cell cDNAs which encode pig antigens that bind to IgG induced after pigto-primate cardiac xenotransplantation. One of these gene products was a glycosyltransferase with homology to the bovine b1,4 N-acetylgalactosaminyltransferase (B4GALNT2). We now characterize the porcine B4GALNT2 gene sequence, genomic organization, expression, and functional significance. Methods: The porcine B4GALNT2 cDNA was recovered from the original library isolate, subcloned, sequenced, and used to identify a bacterial artificial chromosome (BAC) containing the entire B4GALNT2 locus from the Children’s Hospital Oakland Research Institute BACPAC Resource Centre (#AC173453). PCR primers were designed to map the intron/exon genomic organization in the BAC clone. A stable human embryonic kidney (HEK) cell line expressing porcine B4GALNT2 (HEK-B4T) was produced. Expression of porcine B4GALNT2 in HEK-B4T cells was characterized by immune staining and siRNA transfection. The effects of B4GALNT2 expression in HEK-B4T cells was measured by flow cytometry and complement mediated lysis. Antibody binding to HEK and HEK-B4T cells was used to detect an induced antibody response to the B4GALNT2 produced glycan and the results were compared to GTKO PAEC specific non-Gal antibody induction. Expression of porcine B4GALNT2 in pig cells and tissues was measured by qualitative and quantitative real time reverse transcriptase PCR and by Dolichos biflorus agglutinin (DBA) tissue staining. Results: The porcine B4GALNT2 gene shares a conserved genomic organization and encodes an open reading frame with 76 and 70% amino acid identity to the human and murine B4GALNT2 genes, respectively. The B4GALNT2 gene is expressed in porcine endothelial cells and shows a broadly distributed expression pattern. Expression of porcine B4GALNT2 in human HEK cells (HEK-B4T) results in increased binding of antibody to the B4GALNT2 enzyme, and increased reactivity with anti-Sda and DBA. HEK-B4T cells show increased sensitivity to complement mediated lysis when challenged with serum from primates after pig to primate cardiac xenotransplantation. In GTKO and GTKO:CD55 cardiac xenotransplantation recipients there is a significant correlation between the induction of a non-Gal antibody, measured using GTKO PAECs, and the induction of antibodies which preferentially bind to HEK-B4T cells.

Guerard W. Byrne,1,2 Zeji Du,1 Paul Stalboerger,1 Heide Kogelberg2 and Christopher G. A. McGregor1,2 1

Department of Surgery, Mayo Clinic, Rochester, MN, USA, 2Institute of Cardiovascular Science, University College London, London, UK

Key words: antigen – B4GALNT2 – cardiac xenotransplantation – immune response – porcine Abbreviations: B4GALNT2, b1,4 N-acetylgalactosaminyltransferase; BLAST, Basic local alignment search tool; DBA, Dolichos biflorus agglutinin; Gal, galactose a 1,3, galactose; GalNAc, N-acetylgalactosamine; GTKO, pigs with a GGTA-1 a-galactosyltransferase mutation; HEK, human embryonic kidney cells; HEK-B4T, HEK cells expressing the porcine B4GALNT2 gene; HUVEC, Human umbilical vein endothelial cells; Mvwf-1, modifier of von Willebrand factor-1 mutation; Neu5GC, N-acetylneuraminic acid; PAEC, porcine aortic endothelial cell; PBMC, peripheral blood mononuclear cell; RT-PCR, reverse transcriptase dependent polymerase chain reaction; Sda, The SID blood group expressing a GalNAc b1-4[Neu5Ac a2-3]Gal b1-4GlcNAc b1-3Gal terminal glycan. Address reprint requests to Guerard W. Byrne, Departments of Medicine and Surgery, Institute of Cardiovascular Science, Rayne Building Rm. 229, 5 University St., London WC1E 6JF, UK (E-mails: [email protected]; schumacher. [email protected]) This is an open access article under the terms of the Creative Commons Attribution-NonCommercialNoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Received 18 March 2014; Accepted 28 May 2014

543

Byrne et al. Conclusion: The functional isolation of the porcine B4GALNT2 gene from a PAEC expression library, the pattern of B4GALNT2 gene expression and its sensitization of HEK-B4T cells to antibody binding and complement mediated lysis indicates that the enzymatic activity of porcine B4GALNT2 produces a new immunogenic non-Gal glycan which contributes in part to the non-Gal immune response detected after pig-to-baboon cardiac xenotransplantation.

Introduction

Carbohydrate modifications on glycoproteins and glycolipids are involved in a wide array of biological processes including protein stability, development, and cell growth. Variations in carbohydrate modification between individuals and between species define a form of immune self-recognition. Differences in expression of ABH blood group antigens between individuals has long been recognized as a prohibitive boundary across which blood transfusion and organ transplantation is generally not performed. Individuals which do not express a particular ABH antigen are stimulated by intestinal microflora to produce antibody to that glycan. This preformed antibody can induce a hemolytic reaction or promote hyperacute or accelerated allograft rejection of ABH incompatible grafts. The same process occurs across species. Humans and Old World primates do not produce oligosaccharides with terminal galactose a 1,3, galactose (aGal), whereas all other mammals synthesize terminal aGal glycans [1]. As a consequence, humans make high levels of anti-Gal antibody. Anti-Gal antibody is a major immune barrier to xenotransplantation [2]. This antibody has also been implicated in other clinical pathologies, notably the accelerated calcification of glutaraldehyde-fixed porcine bioprosthetic material and degeneration of bioprosthetic replacement heart valves [3,4]. Targeted genetic modification has been used to mutate the a-galactosyltransferase (GGTA-1) gene of the pig (GTKO) [5–7]. This mutation eliminates the enzyme function and, when homozygous, blocks the synthesis of terminal aGal moieties on glycoproteins [8–11]. Porcine GTKO tissue does not bind anti-Gal antibody and transplantation of GTKO cells, tissue and organs does not induce a specific anti-Gal antibody response. Elimination of this antigen has not eliminated GTKO xenograft rejection which remains a predominantly antibody-mediated process now directed to non-Gal antigens [12,13]. We have used proteomic analysis of porcine aortic endothelial cell (PAEC) membranes and expression library screening of PAEC 544

cDNA libraries to identify immunogenic non-Gal target antigens that may contribute to xenograft rejection [14,15]. Using sera, obtained after pig-tobaboon cardiac xenotransplantation, these studies identified 43 potential non-Gal target antigens including six cDNAs, isolated from PAEC expression libraries. When expressed on human embryonic kidney 293 cells (HEK) these cDNAs each produce a porcine antigen that binds to baboon non-Gal IgG. A BLAST search identified one of these recovered cDNAs as a porcine glycosyltransferase with homology to a predicted Bos Taurus b1,4 N-acetylgalactosaminyltransferase (B4GALNT2) sequence [14]. In humans and mice the B4GALNT2 gene catalyzes the terminal addition of N-acetylgalactosamine to a sialic acid modified lactosamine to produce GalNAc b1-4[Neu5Ac a23]Gal b1-4GlcNAc b1-3Gal, the Sda (Sid blood group, also known as CAD or CT) blood group antigen [16,17]. In this study, we further characterize the porcine B4GALNT2 coding sequence, genomic organization, expression, and functional significance.

Materials and methods Cells, culture conditions, and transfection methods

Porcine aortic endothelial cells (PAECs) were isolated from GTKO A- and O-type blood group pig aortas as previously described [18] and cultured in EGM media (Lonza Inc., Allendale, NJ, USA). Porcine peripheral blood mononuclear cells (PBMCs) were isolated from GTKO pig blood by density gradient centrifugation using Ficoll–Hypaque. Human embryonic kidney 293 cells (HEK) and HEK cells expressing the porcine B4GALNT2 cDNA were grown in DMEM media with 10% FBS at 37 °C in a 5% CO2 incubator. The full length pig B4GALNT2 open reading frame was amplified from the original library isolated clone using a primers set containing a Kozak consensus sequence and in-frame translation stop signals, respectively (Forward: 50 - ACCATGACTTCGTA CAGCCCTAG-30 , Reverse: 50 - CAGATACCTTA GGTGGCACATTGGAG-30 ). The PCR product

Porcine B4GALNT2 encodes a new xenoreactive antigen was inserted into pcDNA3.1/V5-His TOPO expression vector (Life Technologies, Paisley, UK) and transfected into HEK cells using Lipofectamine-2000 (Life Technologies). A stable G418 resistant HEK clone expressing the porcine B4GALNT2 genes (HEK-B4T) was established and used for further real-time RT-PCR, immunohistochemical staining, complement dependent cytotoxicity, and flow cytometry analysis. siRNA isolation and transfection

B4GALNT2 specific siRNA and control siRNA for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were generated from the coding region of each gene using a BLOCK-iT Dicer RNAi Kit (Life Technologies) according to the manufacturer’s instruction. RNAi primers for B4GALNT2 were: Forward: 50 -ATGACTTCGTACAGCCCT AG-30 and Reverse: 50 -CAGATACCTTAGGTG GCACATTGGAG-30 . RNAi primers for GAPDH were Forward: 50 -AGGTGAAGGTCGGAGTCA AC-30 and Reverse: 50 -AGTGGTCGTTGAGG GCAATG-30 . The diced siRNA was purified through a RNA Spin Cartridge (Life Technologies). B4GALNT2 and GAPDH diced siRNAs were transfected into HEK-B4T cells using Lipofectamine-2000 (Life Technologies). After 48 h, total RNA was extracted from cells using RNeasy Plus Mini kit (Qiagen Inc., Valencia, CA. USA). B4GALNT2 expression was monitored by real-time RT-PCR and immunohistochemical staining. Real-time and qualitative RT-PCR analysis of B4GALNT2 expression

Total RNA was extracted from cultured cells using RNeasy Plus Mini kit (Qiagen Inc.). Porcine tissue RNA was extracted using 1 g of frozen pig tissue homogenized in 10 ml cold STAT-60 RNA (Tel-Test Inc., Friendswood, TX, USA) and processed as recommended. For both qualitative and quantitative reverse transcriptase PCR (RT-PCR) RNA quality was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies Inc., Santa Clara, CA, USA). To survey the expression of porcine B4GALNT2 a gene specific one-step reverse transcriptase PCR (RT-PCR) assay using 0.5 lg of total tissue RNA and primers for pig B4GALNT2 (Forward: TACAGCCCTAGATGTCTGTC in exon 1, Reverse: CTCTCCTCTGAAAGTGTTCGAG in exon 3) was used to amplify a 334 bp product. Primers for beta-actin (Forward: CAAGATCATCGCGCCTC CA exon 6, Reverse: ACTCCTGCTTGCTGAT CCACATCT, exon 6) produce a 108 bp product

which was used as a loading control. The authenticity of each amplified PCR product was confirmed by sequencing. Gene specific amplification was performed in a MyCycler Thermal Cycler (Bio-Rad, Hercules, CA, USA) using one-step RT-PCR reaction (USB-Affymetrix, Santa Clara, CA, USA) with reverse transcription performed at 50 °C for 30 min. Amplification conditions consisted of 95 °C for 10 min, and 30 cycles of 94 °C for 30 s, 60 °C for 30 s and 72 °C for 30 s. Amplification products were run in a 1.5% agarose gel. For quantitative real-time analysis (Figs 3C and 4B) QuantiTect SYBR Green one-step RT-PCR was used (Qiagen Inc.) according to the manufacture’s instruction. The cycling conditions were as follows: 50 °C for 30 min for reverse transcriptase, 95 °C for 15 min, and 40 cycles of 94 °C for 15 s, 60 °C for 30 s, and 72 °C for 30 s. The B4GALNT2 primers for quantitative real time PCR (Fig. 3C) were: rt-Forward: 50 -GCGACTCCAAAGAATTGGCTTC-30 (exon 10) and rt-Reverse 50 TGGTGACCTATGATCACGTGTG-30 (exon 11) which produces a 120 bp RT-PCR product. A normalized dCT was calculated using b-actin expression and the relative change in B4GALNT2 expression, ddCT, was calculated based on B4GALNT2 expression in the absence of siRNA [19]. Pig and PAEC AO blood group determination was performed as described [20]. Cell staining and flow cytometry

Fluorescein isothiocyanate conjugated Dolichos biflorus agglutinin (FITC-DBA; Vector Labs, Burlingame, CA, USA) and an anti-Sda antibody KM694 (kindly provided by Kyowa Hakko Kirin Co. Ltd., Tokyo, Japan) were used to detect the glycan products of porcine B4GALNT2 expression in cultured human cells. Cells (2.5 9 105) in FACS buffer (phosphate buffered saline (PBS) with 1% bovine serum albumin) were stained with 1 to 5 lg/ml KM694 or 10 lg/ml of FITC-DBA for 45 min at 4 °C. Cells stained with KM694 were washed in FACS buffer and subsequently stained with a FITC conjugated goat anti-mouse IgM (2 lg/ml Southern Biotech, Birmingham, AL, USA). All cells were analyzed using a FACSCalibur (BD Biosciences, San Jose, CA, USA) cytometer and CellQuest software. Immunohistochemical staining of HEK and HEK-B4T cells with rabbit anti-human B4GALNT2 (Sigma-Aldrich, Saint Louis., MO, USA) was used to assess B4GALNT2 protein expression after siRNA transfection. Cells were fixed in cold methanol, washed with PBS containing 0.1% Triton 545

Byrne et al. X-100 and blocked with 10% non-immune serum. Cells were incubated with rabbit anti-human B4GALNT2 in FACS buffer (1 : 75 dilution) for 2 h and washed with PBS. Anti-B4GALNT2 binding was detected with an anti-rabbit diaminobenzidine kit (ZymedLife Technologies, Paisley, UK) as recommended by the manufacturer. Induced antibody directed to the B4GALNT2 produced porcine glycan after pig-to-baboon heterotopic GTKO or GTKO:CD55 cardiac xenotransplantation was measured by comparing differential antibody binding to HEK and HEKB4T cells. Cells (2.5 9 105) were stained with pretransplant, and post-explant serial serum dilutions (1 : 5 to 1 : 40) at 4 °C for 45 min in FACS buffer. After washing, bound antibody was detected using a phycoerythrin conjugated goat anti-human IgM or IgG secondary. Induction of antibody specific for the B4GALNT2 produced glycans was estimated as the ratio of induced antibody binding to HEK-B4T cells compared to HEK cells (antiB4GALNT2 glycan = MFI post explant (HEKB4T)/pretransplant (HEK-B4T): MFI post-explant (HEK)/pretransplant (HEK). Post-explant induction of non-Gal antibody was measured using GTKO PAECs as previously described using serum samples from earlier GTKO or GTKO: CD55 heterotopic cardiac transplants [11] or transplant recipients subject to similar immune suppression. The induced antibody binding data were used to calculate a Spearman rank correlation coefficient for IgM and IgG specific reactivity to HEK-B4T cells and GTKO PAECs. A two tailed non-directed P-value < 0.05 was considered significant. In Table 2, the induced HEK-B4T antibody response is represented in a semi-quantitative scale with ratios