• Aucun résultat trouvé

Identification of the tetraspanin CD82 as a new barrier to xenotransplantation

N/A
N/A
Protected

Academic year: 2022

Partager "Identification of the tetraspanin CD82 as a new barrier to xenotransplantation"

Copied!
11
0
0

Texte intégral

(1)

Article

Reference

Identification of the tetraspanin CD82 as a new barrier to xenotransplantation

SALEH, Soad M, et al.

Abstract

Significant immunological obstacles are to be negotiated before xenotransplantation becomes a clinical reality. An initial rejection of transplanted vascularized xenograft is attributed to Galα1,3Galβ1,4GlcNAc-R (Galα1,3-Gal)-dependent and -independent mechanisms. Hitherto, no receptor molecule has been identified that could account for Galα1,3-Gal-independent rejection. In this study, we identify the tetraspanin CD82 as a receptor molecule for the Galα1,3-Gal-independent mechanism. We demonstrate that, in contrast to human undifferentiated myeloid cell lines, differentiated cell lines are capable of recognizing xenogeneic porcine aortic endothelial cells in a calcium-dependent manner.

Transcriptome-wide analysis to identify the differentially expressed transcripts in these cells revealed that the most likely candidate of the Galα1,3-Gal-independent recognition moiety is the tetraspanin CD82. Abs to CD82 inhibited the calcium response and the subsequent activation invoked by xenogeneic encounter. Our data identify CD82 on innate immune cells as a major "xenogenicity sensor" and open new avenues of intervention to [...]

SALEH, Soad M, et al . Identification of the tetraspanin CD82 as a new barrier to xenotransplantation. Journal of Immunology , 2013, vol. 191, no. 5, p. 2796-805

DOI : 10.4049/jimmunol.1300601 PMID : 23872050

Available at:

http://archive-ouverte.unige.ch/unige:43389

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

The Journal of Immunology

Identification of the Tetraspanin CD82 as a New Barrier to Xenotransplantation

Soad M. Saleh,* Ranjit S. Parhar,* Reem S. Al-Hejailan,* Razan H. Bakheet,*

Hala S. Khaleel,* Hanif G. Khalak,

Anason S. Halees,

Marya Z. Zaidi,*

Brian F. Meyer,

Gisella P. Yung,

x

Jo¨rg D. Seebach,

x

Walter Conca,*

,{,

Khalid S. Khabar,

‡,

Kate S. Collison,* and Futwan A. Al-Mohanna*

,‖

Significant immunological obstacles are to be negotiated before xenotransplantation becomes a clinical reality. An initial rejection of transplanted vascularized xenograft is attributed to Gala1,3Galb1,4GlcNAc-R (Gala1,3-Gal)–dependent and –independent mechanisms. Hitherto, no receptor molecule has been identified that could account for Gal

a

1,3-Gal–independent rejection. In this study, we identify the tetraspanin CD82 as a receptor molecule for the Gal

a

1,3-Gal–independent mechanism. We demonstrate that, in contrast to human undifferentiated myeloid cell lines, differentiated cell lines are capable of recognizing xenogeneic porcine aortic endothelial cells in a calcium-dependent manner. Transcriptome-wide analysis to identify the differentially expressed transcripts in these cells revealed that the most likely candidate of the Gala1,3-Gal–independent recognition moiety is the tetraspanin CD82. Abs to CD82 inhibited the calcium response and the subsequent activation invoked by xenogeneic encounter. Our data identify CD82 on innate immune cells as a major “xenogenicity sensor” and open new avenues of intervention to making xenotransplantation a clinical reality.

The Journal of Immunology, 2013, 191: 2796–2805.

T he acute shortage of donor organs leads to so many deaths of patients in dire need of transplantation. It is estimated that the global number of patients requiring heart trans- plantation is

∼800,000 whereas the total number of hearts trans-

planted in 2007 reached only 3,500 (1). One viable option for donor organ shortage is the use of animal organs as replacements, that is, xenotransplantation. Initially a transplanted organ between discordant species appears viable and healthy, but this is rapidly followed by hyperacute rejection attributed to xenoreactive natural Abs and complement activation (2, 3). Recipient xenoreactive

natural Abs target Gala1,3-Gal, which “decorates” proteins and lipids of the transplanted organ endothelium (4, 5). These deco- rations are brought about by the enzyme

a1,3-galactosyltransfer-

ase (GalT), which is expressed in all mammals except humans, apes, and old world monkeys (6, 7). Many strategies have been employed to overcome hyperacute rejection. These include re- moval of the anti–Gala1,3-Gal Abs (8), accommodation (9), transgenesis (10, 11), and small interfering RNA silencing of the GalT (12). GalT knockout (KO) donor organs gave a glimpse of hope through extending the life of the transplanted organ but succumbed to rejection, eventually albeit at a considerably later time (13, 14). Clinical xenotransplantation is controversial owing to the identified rejection problems and the possibility of xen- ozoonotic diseases (8, 15). Neutrophils and NK cells were iden- tified as Gala1,3-Gal–independent players in xenograft rejection.

We and others have previously demonstrated the xenogeneic recognition and activation of neutrophils and NK cells by porcine aortic endothelial cells (POAECs) in the absence of xenoreactive natural Abs and complement activation through a calcium- dependent mechanism (16–19). The molecular mechanisms un- derlying such Gala1,3-Gal–independent recognition have yet to be determined. In this study, POAECs from wild-type (WT) and GalT KO animals confirm that recognition of xenogeneic endo- thelial cells occurs independently of Gala1,3-Gal. Furthermore, we used three human myeloid cell lines (HL-60, THP-1, and KG- 1) that, in their undifferentiated state, do not recognize xenogeneic endothelial cells as defined by the lack of calcium transients and reactive oxygen metabolite (ROM) production in response to POAECs GalT KO and POAECs WT. However, when differenti- ated, these cells transiently raise their intracellular calcium and increase ROM production upon exposure to either POAECs GalT KO or POAECs WT. To identify possible Gala1,3-Gal–indepen- dent sites mediating the recognition of xenogeneic endothelial cells, we used serial analysis of gene expression (SAGE). SAGE libraries of the myeloid cell lines were used to compare tran- scriptomics before and after differentiation with that in resting

*Department of Cell Biology, King Faisal Specialist Hospital and Research Center, Riyadh 11211, Saudi Arabia;Department of Genetics, King Faisal Specialist Hos- pital and Research Center, Riyadh 11211, Saudi Arabia;Molecular BioMedicine Program, King Faisal Specialist Hospital and Research Center, Riyadh 11211, Saudi Arabia;xUniversity Hospital and Medical Faculty, CH-1211 Geneva 14, Switzerland;{Department of Medicine, King Faisal Specialist Hospital and Research Center, Riyadh 11211, Saudi Arabia; andMedical College, Alfaisal University, Riyadh 11211, Saudi Arabia

Received for publication March 4, 2013. Accepted for publication June 20, 2013.

This work was supported by the Cardiovascular Research Program at King Faisal Specialist Hospital and Research Center and partially by King Abdulaziz City for Science and Technology, National Science, Biotechnology, and Innovation Plan Grant 08-MED487-20.

The sequences presented in this article have been submitted to the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) under accession numbers GSE43211 and GSE1439.

Address correspondence and reprint requests to Prof. Futwan A. Al-Mohanna, De- partment of Cell Biology, King Faisal Specialist Hospital and Research Center, Riyadh 11211, Saudi Arabia. E-mail address: Futwan@kfshrc.edu.sa

The online version of this article contains supplemental material.

Abbreviations used in this article: fluo-3-AM, fluo-3-acetoxymethyl ester; Gala1,3- Gal, Gala1,3Galb1,4GlcNAc-R; GalT,a1,3-galactosyltransferase; HAEC, human aortic endothelial cell; KO, knockout; LDCL, luminol-dependent chemilumines- cence; POAEC, porcine aortic endothelial cell; ROM, reactive oxygen metabolite;

SAGE, serial analysis of gene expression; WT, wild-type.

This article is distributed under The American Association of Immunologists, Inc., Reuse Terms and Conditions for Author Choice articles.

CopyrightÓ2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1300601

(3)

human naive neutrophils. This strategy yielded a number of transcripts that were 1) differentially expressed in all of the dif- ferentiated versus undifferentiated cell lines and 2) constitu- tively expressed in human naive neutrophils. Twelve differentially expressed transcripts were identified by this approach, with only six transcripts displaying consistent change in all three cell lines and in human naive neutrophils. Because the putative xenor- ecognition moieties should be both trans-plasma membrane pro- teins and associated with intracellular calcium release, only one of the six identified transcripts encoding the tetraspanin CD82 met the above criteria and therefore was considered the likely candi- date mediating the Gala1,3-Gal–independent recognition. This was confirmed by subsequent analysis that demonstrated that Abs to CD82 significantly inhibited both the calcium rise and ROM production in human naive neutrophils upon exposure to POAECs GalT KO and POAECs WT. We therefore propose that a CD82- mediated interaction of innate immune cells with xenogeneic endothelial cells is one of the mechanisms employed to recognize interspecies xenogenicity.

Materials and Methods

Materials

Fluo-3-acetoxymethyl ester (fluo-3-AM) and fura 2-AM were purchased from Molecular Probes (Invitrogen, Carlsbad, CA). LightCycler DNA Master SYBR Green I was purchased from Roche Diagnostics (Mannheim, Germany). An I-SAGE/I-Long SAGE kit with magnetic stand, Platinum Taq DNA polymerase, and TRIzol solution were purchased from (Invitrogen).

Cell lines were purchased from the American Type Culture Collection (Rockville, MD). Culture media (RPMI 1640 and DMEM) were purchased from Life Technologies BRL (Grand Island, NY). Abs to von Willebrand factor were purchased from Sigma-Aldrich (St. Louis, MO), and acetylated low-density lipoprotein (DiI-Ac-LDL) was purchased from Biogenesis (Bournemouth U.K.). Abs to CD82 were purchased from Abcam (Cam- bridge, MA) and Santa Cruz Biotechnology (Santa Cruz, CA). Anti–LFA- 1awas purchased from R&D Systems (Minneapolis, MN). Anti–Gala1,3- Gal Abs were prepared essentially as described previously (9). Secondary goat anti-mouse FITC-labeled Abs were purchased from Santa Cruz Biotechnology and Sigma-Aldrich, and Alexa Fluor 647–labeled second- ary Abs were purchased from Pierce. PMA, DMSO, dibutyryl cAMP, MTT, cell culture reagents, protease inhibitors, and other analytical-grade reagents were purchased from Sigma-Aldrich. Restriction enzymes NlaIII, MmeI, and Sph were purchased from New England Biolabs (Beverly, MA). Fluo-3-AM, Fura-2-AM, and luminol were dissolved in DMSO and delivered to the cells at a final concentration of 1, 1, and 11mM, respec- tively, in a final DMSO concentration of 0.1%.

Preparation of endothelial cells

POAECs (P304K-05) and human aortic endothelial cells (HAECs; 304K- 05a) were purchased from Cell Applications (San Diego, CA). HUVECs (CC-2517) were purchased from the Lonza Group (Basel, Switzerland).

POAECs and HAECs were cultured and maintained in tissue culture medium from Cell Applications, whereas HUVECs were cultured and maintained in tissue culture medium purchased from Life Technologies.

Cells were used from passages 2 through 10 in all experiments at a split ratio of 1:3. Additionally, aortic endothelial cells isolated from WT pigs and POAECs derived from GalT KO pigs were used when appropriate.

These cells were provided by R. J. Hawley (formerly of Immerge Bio- Therapeutics, Cambridge, MA). Cell lines were obtained and cultured as previously reported (20–22). To test that endothelial cells were not ac- tivated during culture, IL-1 levels in conditioned medium were measured using ELISA (R&D Systems) and were consistently found to be negli- gible (,4 pg/ml).

Preparation of neutrophils

Human peripheral blood neutrophils were prepared by dextran sedimen- tation of heparinized whole blood obtained from healthy donors and centrifuged through Ficoll-Paque as described previously (16). Neutrophils were routinely tested for production of ROMs by luminol-dependent chemiluminescence (LDCL) for 10 min. Cells were considered naive and therefore suitable for experimentation only when no increase in LDCL was observed.

Preparation of cell Lines HL-60, KG-1, and THP-1

Acute promyelocytic leukemia HL-60 cell line (ATCC CCL-240), acute myelogenous leukemia KG-1 cell line (ATCC CCL-246), and acute monocytic leukemia THP-1 cell line (ATCC TIB-202) were purchased from the American Type Culture Collection. HL-60 and KG-1 cell lines were cultured in complete Iscove’s modified medium (American Type Culture Collection, catalog no. 30-2005) supplemented with 10% FBS (American Type Culture Collection, catalog no. 30-2020), penicillin (100 U/ml), and streptomycin (100mg/ml). THP-1 cells were cultured in complete RPMI 1640 medium (American Type Culture Collection, catalog no. 30-2001) supplemented with 10% FBS (American Type Culture Collection, catalog no. 30-2020), penicillin (100 U/ml), and streptomycin (100mg/ml). All cell lines were maintained in a humidified incubator at 37˚C with 5% CO2. HL-60 differentiation into neutrophil-like cells was performed by treat- ment of 23106cells/ml with 1.3% DMSO (Sigma-Aldrich, catalog no.

D4540) in complete media for 6–8 d with media change every third day.

Differentiation into neutrophil-like cells was ascertained by their ability to generate ROMs in response to stimulation by PMA (100 ng/ml) or the chemotactic peptide fMLF (1mM). This was detected by either the re- duction of the soluble NBT to blue-black insoluble formazan and/or LDCL. For the former, 1 ml cell suspension was incubated for 20 min at 37˚C with an equal volume of 0.2% NBT (Sigma-Aldrich) dissolved in PBS (pH 7.2; 0.15 M without Ca2+, Mg2+) in the presence of 200 ng PMA.

Differentiated cells contain formazan deposits as dark, irregularly shaped crystal inclusions in the cytoplasm. By day 6,∼98% of the cells reduced NBT upon PMA stimulation and,5% of the cells reduced NBT in the absence of PMA stimulation. THP-1 and KG-1 differentiation was per- formed as above but with treatment with dibutyryl cAMP (500mg/ml) and PMA (100 ng/ml) for 4 and 5 d, respectively (23). Differentiation was confirmed by ROM production as above.

Calcium measurements

Neutrophils or appropriate cell lines were loaded with fluo-3-AM (1mM) as described previously (18). The cells were washed, placed on glass cover- slips, and allowed to adhere for 15 min at room temperature. Coverslips were washed and then secured between two plates of a custom-designed coverslip holder, placed onto a heated microscope stage (37˚C), and in- tracellular Ca2+concentration images were acquired at 1- to 2-s intervals.

For each coverslip, 100 ml POAECs WT and POAECs GalT KO, sus- pended in Krebs-HEPES buffer (pH 7.4) and containing 104cells, were added and image acquisition was continued for at least 5 min. Control experiments were carried out using equal numbers of HAECs or HUVECs per coverslip. Images were analyzed using UltraVIEW confocal software (PerkinElmer, Cambridge, U.K.) and fluorescence intensity (from each cell) was transformed into absolute calcium levels as described previously (16). Because undifferentiated cell lines were nonadherent, calcium measurements were carried out using fluorimetric assays (PerkinElmer LS 55 luminescence spectrometer) with cells labeled with fura 2-AM as described previously (24). Experiments were analyzed using FL WinLab software (Per- kinElmer). In experiments involving anti-CD82 Abs we used mouse mono- clonal anti-human CD82 (TS82b, ab59509, lot GR 25852-1; Abcam).

Measurement of ROM production

The production of ROMs by neutrophils and the three cell lines were measured using LDCL on an FB12 single tube luminometer (Berthold Detection Systems/Titertek Instruments, Huntsville, AL), essentially as described previously (18).

RNA isolation

Total cellular RNA was isolated from 1 to 2 3107 differentiated/

undifferentiated cell lines and human naive neutrophils using TRI Re- agent (Molecular Research Center, Cincinnati, OH) following the manu- facturer’s instructions. RNA integrity was routinely checked using 500 ng/

ml RNA on 1% denaturing agarose gel.

Construction of 59 Long SAGE libraries

SAGE was performed according to the SAGE detailed protocol version C and analyzed using SAGE analysis software version 4.5 (Johns Hopkins University, Baltimore, MD). In brief, 10mg total RNA was bound to solid- phase oligo(dT) magnetic beads, and cDNA was synthesized directly on the oligo(dT) bead. Oligo(dT) bound to magnetic beads was used as a template for the first-strand cDNA synthesis, followed by the second- strand cDNA synthesis. The captured cDNAs were then digested with an

“anchoring” restriction enzyme, NlaIII, which left a 39overhang. Com- plementary cDNA synthesis and NlaIII digestion was verified using PCR.

The 39fragments were then isolated using magnetic beads, equally divided

(4)

into two pools and ligated to two different linkers, A or B. Both linkers contain the recognition sequence for a “tagging” restriction enzyme (type IIs restriction enzyme), Mme. The tagging enzyme produced a staggered cut, offset by∼17 bp from the recognition sequence. The two linkers were ligated onto the NlaIII overhangs. The efficiency of ligation was assessed by PCR. Subsequent digestion with Mme released the adapter with a short tag of cDNA from the beads. These tags were then ligated tail-to-tail to form 130-bp ditags. The resulting ditags were PCR amplified using primers specific to each linker, pooled, precipitated, and gel purified. The linkers were released by digesting with NlaIII, and the resulting 34-bp ditags were gel purified, concatenated, and resolved on 8% (w/v) polyacrylamide gel.

The high-molecular mass bands (300–500 bp, 500–800 bp, and 800 bp–1 kb) were gel purified and cloned into Sph1-linearized pZErO-1 vector (Invitrogen). Ligation products were transformed into One Shot TOP10 electrocompetent cells. Transformants were analyzed by colony PCR.

Approximately 4000–5000 clones for each library were cycle sequenced using M13 forward primer and analyzed on an Applied Biosystems DNA sequencer. Each concatemer insert results in a randomly organized “series” of ditags of∼34 bp, with each being flanked by the recognition sequence of the primary anchoring enzyme NlaIII CATG sequence that provides a “SAGE tag” specific to each expressed gene. Approximately 20–25 individual tags were produced per clone. SAGE software was used to convert these sequences into Long SAGE tags and tabulate tag abundances. Resultant SAGE tags were analyzed using the downloadable reference sequence database SAGEmap (25) from the National Center for Biotechnology Information Web site. By deter- mining the frequency distribution of the total tag population, the statistical breakdown of the relative abundance of the different mRNAs expressed in the differentiated versus undifferentiated cell populations was obtained. Using this method we generated five libraries, including human naive neutrophils and differentiated and undifferentiated HL-60 and KG-1 cell lines (SAGE data were submitted to the Gene Expression Omnibus repository under ac- cession no. GSE43211; http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?

token=lzyffgkqokkgchc&acc=GSE43211). The differential expression results for THP-1 cells were from the Gene Expression Omnibus data repository at the National Center for Biotechnology Information (accession no. GSE1439).

All SAGE tags identified for the three cell lines were aggregated for each gene across multiple tags. Differential expression significance for these aggregated counts was determined using aztest, wherez= (xd2xu)/v(xd+xu), andpvalues were assigned as,0.01 forz.2.58 and as,0.05 for z.1.96, according to a univariate normal distribution. Subsequently, differentially expressed genes in all three cell lines (pvalues,0.05) were considered.

Quantification of specific transcripts with LightCycler RT-PCR

RNA isolation.Total RNA was extracted from tissue culture cells using an RNeasy kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions. RNA quality was verified using formaldehyde gel. RNA con- centrations were determined by absorption at 260-nm wavelength with an ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE).

Real-time RT-PCR.Total RNA (2mg) was reverse transcribed using a RT system from Promega (Madison, WI). The cDNA was then used as tem- plate for real-time PCR with gene-specific primers.

Gene-specific primers corresponding to the PCR targets were designed using Oligo 6 primer analysis (Applied Biosystems, Foster City, CA). The following oligonucleotides were selected: CD82 splice variant V1 (ac- cession NM_002231), sense, 59-GGTCCTGTCCATCTGCTTGT-39; anti- sense, 59-CCAGAAAGCCCCCTACTTTC-39; CD82 splice variant V2 (accession NM_001024844), sense, 59-GATGGTCCTGTCCATCTGCT- 39, antisense, 59-CCAGAAAGCCCCCTACTTTC-39; human GAPDH (accession NM_002046), sense, 59-GGTGAAGGTCGGAGTCAAC-39, antisense, 59-ATGGGTGGAATCATATTGGA-39; POAECs GalT (acces- sion NM_L36535), sense, 59-CAGTGGTATGGGAAGGCACT-39, anti- sense, 59-AGATGACTT TGTGGCCAACC-39; and POAEC GAPDH (accession NM_001206359), sense, 59-GTCGGTTGTGGATCTGACCT- 39, antisense, 59- AGCTTGACGAAGTGGTCG TT-39;

Conditions for real-time PCRs were first optimized using a gradient cycler (Bio-Rad Laboratories, Foster City, CA). RT-PCR amplification product were separated on 2% agarose gel and analyzed with the Quantity One 1-D software (Bio-Rad Laboratories). Optimized conditions were transferred to the LightCycler real-time PCR protocol. MgCl2 concen- trations were reoptimized using the LightCycler protocol (Roche Molec- ular Biochemicals, Mannheim, Germany). Preliminary real-time RT-PCR experiments were performed with each primer pair to determine the annealing temperature that yielded the greatest amount of specific product with melting temperature separable from primer-dimer temperature.

Standard curves were prepared for each run using known quantities of cDNA (10-fold dilutions beginning at 15 ng/ml) and primers for the gene of interest or GAPDH. LightCycler FastStart DNA Master SYBR

Green I reaction mix containing 0.4mM forward and reverse primers, 4 M MgCl2, and 1ml LightCycler FastStart DNA Master SYBR Green 1 dye was used for quantitative real-time PCR analysis according to the manufacturer’s protocols (Roche Molecular Biochemicals). A volume of 1ml from a 10-fold dilution of cDNA was added as the PCR template.

A no-target control received 9ml reaction mix and 1ml water. PCR amplification was performed in triplicate wells. A four-step experi- mental run protocol was performed: 1) denaturation program (10 min at 95˚C);

2) 45 cycles of four-segment amplifications consisting of denaturation at 95˚C for 10 s, annealing (60–56˚C) for 5 s, extension at 72˚C for 10 s, and data acquisition at 83˚C for 1 s (a temperature transition rate of 2˚C/s with a single fluorescence measurement was used); 3) melting curve program (50–95˚C, with heating rate of 0.1˚C/s up to 98˚C with continuous fluorescence measurement); and 4) a cooling program down to 40˚C. Relative quantitation measurements were performed using external standard curves for both target and GAPDH housekeeping gene. The second derivative maximum method was used for cycle threshold calculation from amplification curves. The respective concentration for any given sample was calculated using crossing-cycle analysis provided by the LightCycler software. For realistic quantifications, the starting amount of RNA was the same for all samples. Minor sampling errors were avoided by normalization against the housekeeping gene GAPDH.

Immunofluorescence and confocal microscopy

Detection of Gala1,3-Gal expression on POAECs was performed using baboon anti–Gala1,3-Gal Abs prepared from emulsified soluble Gala1,3-Gal with Hunter’s TiterMax adjuvant as described previously (9, 19). Briefly, POAECs WT or POAECs GalT KO (13106cells/ml) seeded∼18 h previously on glass coverslips were fixed in 3.7% (v/v) formaldehyde (15 min at room temperature). Cells were washed thor- oughly and Gala1,3-Gal structures were visualized by indirect immu- nofluorescence using baboon anti–Gala1,3-Gal IgG (1:250) and FITC- conjugated goat anti-human secondary Ab (1:500; Sigma-Aldrich). Cells were viewed by laser scanning confocal microscopy (Zeiss LSM 510 META system; Zeiss, Jena, Germany). CD82 and LFA-1adouble labeling was performed on live cells on ice. Briefly, 50ml cell suspension (107cells/ml) was incubated with mouse monoclonal anti-human CD82 Ab (Abcam, lot 85961-0) and goat polyclonal anti-human rLFA-1a (R&D Systems, lot AF1730) at 1:50 and 1:500 dilutions, respectively, for 1 h on ice. Cells were washed and treated with donkey anti-goat Alexa Fluor 647 secondary Ab (Molecular Probes, lot 93D1-1) for 30 min on ice at 1:500 dilution. The cells were washed thoroughly and then treated with goat anti-mouse FITC-labeled secondary Ab (Abcam, ab6785, lot 906955) at 1:500 dilution on ice for 30 min. The cells were then washed and spotted on the center of a coverslip, which was sandwiched between two plates of a specially designed holder, and viewed using a Zeiss LSM 510 META confocal microscope.

Flow cytometry

POAECs WT or POAECs GalT KO (13106cells/ml) were labeled with anti–Gala1,3-Gal Abs (1:250) for 45 min on ice. The cells were washed three times (PBS [pH 7.2]) and then labeled with secondary Ab (FITC- conjugated goat anti-human IgG, 1:500; BD Biosciences, San Jose, CA, catalog no. 555988) for another 45 min on ice. Isotype IgG (IgG1/IgG isotype; BD Biosciences, catalog no. 349526) or secondary Ab alone was used as a control. Finally, the cells were washed with PBS, fixed in 1%

paraformaldehyde, and analyzed by a BD FACSCalibur (BD Biosciences).

Ten thousand cells were analyzed in each sample to express the percentage of positive cells.

Sample preparation for Western blots

Appropriate cells (23106cells/ml) were treated with 1.5 ml RIPA buffer (150 mM NaCl, 1.0% Nonidet P-40 or 0.1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl [pH 8.0], and protease inhib- itors), scraped, transferred into a precooled centrifuge tube, and homoge- nized on ice. The homogenized cells were centrifuged at 16,0003gfor 20 min, supernatants were collected, and protein concentrations were deter- mined. Samples were run in 4–12% SDS-PAGE gel, transferred onto ni- trocellulose membranes, and washed thoroughly. Membranes were blocked (5% nonfat milk in TBST buffer) for 1 h and then probed with primary rabbit anti-human CD82 (H-173, Sc-5540; (Santa Cruz Biotechnology) at a 1:200 dilution in blocking buffer overnight at 4˚C. Membranes were washed three times (5 min each) with TBST and incubated in sec- ondary anti-rabbit HRP-labeled Ab (W401B, 24947003; Promega) diluted 1:2500 with blocking buffer for 1 h at room temperature.

Membranes were wash five times with TBST (10 min each) and probed by ECL (Amersham Pharmacia Biotech, Little Chalfont, U.K.).

2798 ROLE OF CD82 IN INTERCELLULAR XENOGENEIC MOLECULAR DIALOGUE

(5)

Results

Calcium-dependent recognition of xenogeneic endothelial cells by human naive neutrophils is independent of Gala1,3-Gal The absence of Gala1,3-Gal epitope in POAECs GalT KO was confirmed by confocal laser scanning microscopy and flow

cytometry, which showed negative staining with baboon anti–

Gala1,3-Gal-labeled POAECs GalT KO compared with POAECs WT (Fig. 1A). Further confirmation was obtained by the total absence of the enzyme GalT in POAECs GalT KO but not in POAECs WT (Fig. 1A). Exposure of human naive neutrophils to

FIGURE 1. (A) Fluorescence confocal micrographs of POAECs WT (upper top panel,left) and POAECs GalT KO (lower top panel,left) labeled with baboon anti–Gala1,3-Gal IgG prepared as described inMaterials and Methods. The corresponding bright field micrographs (upperandlower top panel, center) and the superimposed images (upperandlower top panel, right). Themiddle panelshows flow cytometry data obtained from POAECs WT and POAECs GalT KO labeled with baboon anti–Gala1,3-Gal IgG, demonstrating Gala1,3-Gal–labeled POAECs WT (upper right quadrant) compared with lack of label in POAECs GalT KO cells (lower right quadrant). Corresponding control experiments with preimmune IgG fraction from baboon sera are shown on theupperandlower left quadrants, respectively. Thebottom panelshows the expression of the GalT gene in POAECs WT but not in POAECs GalT KO cells using RT-PCR. Data are expressed as the GalT/GAPDH ratio. Experiments were performed as described inMaterials and Methods. Data are representative of at least three independent experiments. (B) Calcium maps following exposure of human naive neutrophils to POAECs GalT KO (104/ml,top panel) at the indicated intervals. Thebottom panelshows absolute intracellular calcium levels following xenogeneic exposure. The arrow indicates the time of addition of POAECs GalT KO. Data are representative of at least three independent experiments. Calcium levels are expressed as means6SEM (n= 53 cells). (C) Calcium maps in human naive neutrophils after exposure to POAECs WT (104/ml) (top right) and in the presence of saturating concentration of anti–Gala1,3- Gal IgG (top left). Calcium changes are color coded such that high calcium is indicated by warm colors.Bottom panelsindicate the absolute calcium levels after exposure. Arrow indicates the times of addition of POAECs WT. Data are expressed as means6SEM (n= 26 and 46 cells, respectively).

(6)

POAECs GalT KO (2

3

10

6

/ml) resulted in a rise in their intra- cellular Ca

2+

concentrations from the resting level of 69.5

6

0.15 to 421

6

20 nM before decaying back to pre–POAECs GalT KO encounter (Fig. 1A, 1B). This rise was largely dependent on re- lease from intracellular stores, because parallel experiments per- formed in calcium-free medium in the presence of extracellular EGTA (1 mM) exhibited no significant difference in the extent of POAECs GalT KO–induced calcium rise. The calcium response was always asynchronous and heterogeneous. Similar results were obtained by POAECs WT where the neutrophil calcium rose to 499

6

33 nM from the resting level of 70

6

0.1 nM upon xeno- geneic encounter (Fig. 1C). The calcium transient was unaffected by the presence of saturating concentrations of anti–Gala1,3-Gal Abs or the absence of xenoreactive natural Abs and complement.

Moreover, neither HAECs nor HUVECs evoked any calcium rise in human naive neutrophils (16).

Activation of human naive neutrophils by POAECs GalT KO and POAECs WT but not by allogeneic endothelial cells Activation of human naive neutrophils following xenogeneic en- counter was tested by measuring ROM production using LDCL. In a series of experiments we found that POAECs GalT KO and POAECs WT elicited ROM production in human naive neutrophils (Fig. 2A). In contrast, neither HAECs nor HUVECs exhibited any effects on ROM production (Fig. 2A). Parallel experiments in the presence of satu- rating concentrations of Abs to Gala1,3-Gal failed to affect ROM production by either POAECs GalT KO and POAECs WT (Fig. 2B).

Undifferentiated human cell lines HL-60, THP-1, and KG-1 do not recognize xenogeneic endothelium unless differentiated into neutrophil-like or monocyte-like cells

The question arose as to the identity of the Gala1,3-Gal–inde- pendent sites mediating the recognition of xenogeneic endothelial

cells by human naive neutrophils. Because both POAECs GalT KO and POAECs WT xenogeneic endothelial cells were recog- nized by human naive neutrophils, and because the latter are terminally differentiated and therefore should possess all of the necessary components for such recognition, we investigated the ability of undifferentiated human myeloid cell lines to recognize xenogeneic endothelial cells. We tested three cell lines: HL-60 and THP-1 cells, which differentiate into neutrophil-like cells, and KG-1 cells, which differentiate into monocyte-like cells (26–28).

The promyeloblast cell line HL-60 assumes a nonadherent spher- ical morphology that changes into a neutrophil-like phenotype with adhesion capabilities upon differentiation in 1.25% DMSO for 7 d. Using a ratiometric calcium dye, we found that POAECs WT were unable to evoke a significant calcium rise in the HL-60 cells (Fig. 3A). Additionally, these undifferentiated cells failed to pro- duce ROMs upon exposure to the xenogeneic endothelial cells. In contrast, differentiated HL-60 cells displayed a transient calcium rise, with levels reaching 274

6

3 nM from a resting level of 70

6

0.01 nM after exposure to POAECs WT (Fig. 3A), and they were concomitant with ROM production (Supplemental Fig. 3). This suggested that differentiation was a prerequisite for the xenor- ecognition. This was further confirmed by the results obtained with THP-1 cells, which upon treatment with dibutyryl cAMP differ- entiate into neutrophil-like cells (Fig. 3B), and KG-1 cells, which upon treatment with PMA differentiate into monocyte-like cells (Fig. 3C).

Identification of the Gala1,3-Gal–independent xenogeneic recognition moieties by transcriptomics

Because only differentiated cell lines and human naive neutrophils were able to recognize and respond to encounters with xenogeneic POAECs WT, the possibility existed that common molecular moieties in the four cell types may be responsible for this rec- ognition. We therefore used SAGE to identify differentially expressed transcripts in the three cell lines and in human naive neutrophils. We used a snapshot approach of mRNA transcripts of undifferentiated and differentiated HL-60 and KG-1 cells, within which our recognition moieties were expected to exist. Undiffer- entiated HL-60 cells exhibited 14,578 transcripts after sequencing 20,261 tags, whereas differentiated HL-60 cells exhibited 16,277 transcripts following 18,206 sequenced tags. Of those transcripts, 248 were significantly differentially expressed (p

#

0.05) (Fig. 4).

Similarly, undifferentiated KG-1 cells exhibited 31,311 transcripts obtained from 38,793 sequenced tags compared with 22,084 transcripts obtained from 29,810 sequenced tags in the differen- tiated KG-1 cells. Of all transcripts in KG-1 cells (undifferentiated and differentiated), 651 were significantly differentially expressed (Fig. 4). The differentially expressed transcripts from both cell types were then compared with differentially expressed transcripts of the THP-1 cell line available from the Gene Expression Om- nibus data repository at the National Center for Biotechnology Information (accession no. GSE1439). This approach identified 12 differentially and significantly expressed transcripts (p

#

0.05) common to all three cell lines as they differentiate into neutro- phil-/monocyte-like cells. Six transcripts displayed levels of ex- pression that were not consistent in all three cell lines and were therefore excluded (Fig. 4, Table I), leaving six differentially expressed transcripts that were consistently upregulated in the three cell lines and were also expressed in the human naive neutrophils. Because our targets were expected to be associated with the plasma membrane, we used the Database for Anno- tation, Visualization and Integrated Discovery version 6.7 (http://david.abcc.ncifcrf.gov/), GOstat software and goa_hu-

FIGURE 2. Activation of human naive neutrophils by POAECs WT and

POAECs GalT KO. (A) LDCL in human naive neutrophils after encounter with POAECs WT (red), POAECs GalT KO (blue), and HAECs (green). (B) POAEC WT–induced LDCL in human naive neutrophils in the presence (red) and absence (blue) of anti-Gal F(ab9)2. Data are representative of at least five independent experiments with neutrophils isolated from different donors.

2800 ROLE OF CD82 IN INTERCELLULAR XENOGENEIC MOLECULAR DIALOGUE

(7)

man database (http://gostat.wehi.edu.au/), Ingenuity Pathway Analysis (https://analysis.ingenuity.com/pa/public/security.jsp), and Pathway Studio (http://www.ariadnegenomics.com/support/pathway- studio-8/) to analyze the cellular locations of the six transcripts identified above. Five of these transcripts, namely ferritin L chain, ferritin H chain,

g-actin, creatine kinase, and adenylate

cyclase–associated protein-1, were all assigned a cytoplasmic location and were therefore excluded. This left the tetraspanin CD82 as the only differentially expressed trans-plasma mem- brane protein that is associated with the ability of differentiated cell lines and human naive neutrophils to recognize xenogeneic endothelial cells independently of Gala1,3-Gal (Fig. 4).

Confirmation of SAGE results at the mRNA and protein levels To confirm the differential expression of CD82 transcripts we used quantitative RT-PCR and Western blot analysis on samples from undifferentiated and differentiated cell lines and human naive neutrophils. We found that in the undifferentiated HL-60 cells, the ratio of CD82 mRNA transcript relative to GAPDH increased from 2.40

6

0.03 to 20.74

6

0.13 upon differentiation. Similar results

were obtained with the other two cell lines, KG-1 and THP-1 (Fig.

5A). These changes in the message levels were echoed by re- spective increased protein levels in Western blot experiments (Fig.

5A). Localization of the expressed CD82 was examined by con- focal microscopy of live human naive neutrophils and found to be associated with the plasma membrane (Fig. 5B). Further confir- mation of this location was provided by colocalization of CD82 with the adhesion molecule LFA-1a in dual-labeled live neutro- phil experiments (Fig. 5C).

Inhibition of xenogeneic recognition by anti-CD82 Abs

Because SAGE, quantitative RT-PCR, Western blots, and confocal data identified CD82 as the likely candidate for xenogeneic recog- nition, we argued that blocking CD82 should inhibit recognition of POAECs WT by human naive neutrophils. In a series of experiments neutrophils were exposed to anti-CD82 Abs (1

mg/ml for 15 min at

room temperature) prior to xenogeneic contact. This treatment significantly reduced POAECs WT–induced calcium from 482

6

24 to 183

6

12 nM (p

#

0.0001, Fig. 6A–C, Supplemental Figs. 1, 2) with a concomitant decrease in ROM production (Fig. 6D).

FIGURE 3. POAEC WT–induced calcium changes in undifferentiated and differentiated cell lines. Differentiation is indicated by the reduction of soluble NBT to blue-black insoluble formazan. Bright field micrographs of undifferentiated and differentiated cells (upper panel) with the corresponding calcium changes (lower panel) in (A) HL-60, (B) THP-1, and (C) KG-1 cell lines. Red arrows indicate the time of addition of POAECs WT (104cells/ml). Calcium measurements and staining were performed as indicated inMaterials and Methods. Original magnification340.

(8)

Discussion

We and others have previously reported the activation of human naive neutrophils by POAECs in the absence of xenoreactive natural Abs and complement and under conditions where the Gala1,3-Gal binding sites were blocked (16–19). In the present study we demonstrate that human naive neutrophils are activated not only by xenogeneic POAECs WT but also by POAECs GalT KO. Allogeneic HAECs or HUVECs, alternatively, do not invoke such activation. This suggests that the recognition of xenogeneic endothelial cells by human naive neutrophils occurs in a Gala1,3- Gal–independent fashion. The presence of anti–non-Gal xenor- eactive Abs in humans and nonhuman primates suggests the in- volvement of non-Gal recognition sites in the rejection process.

Non-Gal Ags have been implicated in acute humoral xenograft rejection (29). Moreover, Saethre et al. (30) identified anti- Hanganutziu–Deicher Abs and others targeting the non-Gal binding sites on the xenogeneic endothelial cells isolated from POAECs GalT KO animals. The question, however, remains as to whether these non-Gal epitopes are involved in the xenor- ecognition and activation of human naive neutrophils by POAECs.

To gain insight into the molecular moieties involved in this non- Gal–dependent recognition/activation mechanism, we used mye- loid cell lines that in their undifferentiated state are not activated by xenogeneic endothelial cells and only become activated after

differentiation into neutrophil-like or monocyte-like cells. SAGE analysis of these cells and of resting human naive neutrophils revealed six transcripts that were consistently expressed in the differentiated cell lines and thus are the likely candidates re- sponsible for xenorecognition and subsequent activation of these cell lines. Of these six transcripts, CD82 was identified as the likely candidate responsible for xenoendothelial cell recognition.

Three lines of evidence support such a claim: 1) undifferentiated myeloid cell lines expressing relatively low levels of CD82 do not evoke a calcium transient or ROM production in response to POAECs WT or POAECs GalT KO; 2) differentiated myeloid cell lines and human naive neutrophils expressing relatively higher levels of CD82 respond to xenogeneic POAECs WT and POAECs GalT KO; and 3) Abs to CD82 inhibit both calcium transients and ROM production in response to xenogeneic contact.

CD82, also known as C33 Ag or KAI1 and originally identified as a marker for activation/differentiation of mononuclear cells (31), is a member of the tetraspanin family of proteins responsible for divergent cellular activities, including activation (32), differenti- ation (33), motility (34), adhesion (35), signaling (36), fusion (37), and metastasis (38). They are highly conserved and can be found in species as disparate as fungi and mammals (39). In humans, CD82 is expressed in many cell types, including lymphocytes (40), granulocytes (41), epithelial cells (42), platelets (43), en- dothelial cells (44), and fibroblasts (45). Thirty-four mammalian tetraspanins were identified, with 33 expressed in humans (39).

All have four transmembrane domains with cytosolic N- and C- terminal regions and two extracellular domains with a conserved CCG motif (39). CD82 exists as two isoforms resulting from two distinct splice variants. It is heavily palmitoylated and glycated (46) and together with other proteins constitutes the tetraspanin web (47). The web is a complex entity invoking functional di- versity in stimulus-response coupling. Furthermore, CD82 was demonstrated to link lipid rafts to the actin cytoskeleton, and depletion of cholesterol seems to inhibit CD82-dependent re- sponses (48). Tetraspanins have long been considered as mem- brane organizers (48) and are known to interact with a number of integrins (49), growth factors (50), lipid rafts (51), actin cyto- skeleton, and membrane domains (52), suggesting their involve- ment in immune responses and immune synapse (53). Immune synapse, or “supramolecular activation clusters,” is best described in T cell/APC interactions (54) and T cell priming (55). Immune synapse is a major intercellular contact milieu where different proteins cluster to assemble to perform a variety of functions (54).

FIGURE 4. Venn diagram of the differentially expressed transcripts in the three cell lines upon differentiation showing the common consistently expressed genes (p, 0.05). The heat map identifies the differentially expressed genes and their relative expression ratios as indicated by the colored bar.

Table I. Transcriptome analysis using SAGE libraries

Gene Symbol Description SAGE Tags

Differentiated/

Undifferentiated Cells

Neutrophil Counts

HL-60 KG-1 THP-1

CD82 CD82 molecule GTGAAACCCCGTCTCTA 1.63 1.55 1.50 58.81

FTL Ferritin, light polypeptide CCCTGGGTTCTGCCCGC 3.07 1.59 1.65 255.07

CKB Creatine kinase, brain GTGACCACGGGTGACGG 1.55 1.49 1.73 83.68

FTH1 Ferritin, heavy polypeptide 1 TTGGGGTTTCCTTTACC 2.59 3.95 1.62 517.23

ACTG1 Actin,g1 CTAGCCTCACGAAACTG 1.70 1.43 1.40 38.48

CAP1 Adenylate cyclase–associated protein 1 CTCATCAGCTTATGGCT 1.96 2.16 1.82 77.11 TPT1 Tumor protein, translationally controlled 1 TAGGTTGTCTAAAAATA 1.59 22.52 1.50 49.53

GAPDH GAPDH TACCATCAATAAAGTAC 1.41 21.46 22.10 42.99

RPS20 Ribosomal protein S20 GCTTTTAAGGATACCGG 21.87 21.20 1.40 33.64

RPL38 Ribosomal protein L38 GCGACGAGGCGCGCTGG 21.90 21.42 1.64 30

RPS3A Ribosomal protein S3A GTGAAGGCAGTAGTTCT 21.52 22.24 1.50 31.72

RPL34 Ribosomal protein L34 TGTGCTAAATGTGTTCG 21.81 22.06 1.50 17

The analysis revealed 12 differentially expressed transcripts. Only six transcripts exhibited a consistent upward trend in the three cell lines upon differentiation and are present in the human naive neutrophils. Positive change is noted by boldface and negative change is noted by italics.

2802 ROLE OF CD82 IN INTERCELLULAR XENOGENEIC MOLECULAR DIALOGUE

(9)

Whether such immune synapse occurs as part of the xenogeneic recognition mechanism is yet to be determined. It is tempting to speculate that the organizing capability of tetraspanins may direct the assembly of an innate immune synapse at the xenogeneic recognition milieu and the corresponding binding site on the membrane of the xenogeneic endothelial cell, thus initiating in- tercellular crosstalk through mobilization of intracellular calcium in both cell types. Intracellular calcium transients have been demonstrated in POAECs in response to xenogeneic encounter with human naive neutrophils and NK cells (16, 19). Our present data suggest a role for tetraspanins as xenogeneic sensors analo- gous to the role of TCR/MHC class II in allogeneic recognition.

Calcium involvement in tetraspanin signaling has been dem- onstrated through the use of Abs to the various tetraspanins. In platelets, mAbs to CD9 induced calcium transients through Fcg II receptors (56). In contrast in human monocytes and B cells, cal- cium signaling was thought to occur via direct crosslinking of F (ab9)

2

fragments of anti-CD53 Abs (57). Lebel-Binay et al. (31) have demonstrated that anti-CD82 Abs evoked calcium changes in the monocytic cell line U937 through dual binding to CD82 and Fcg receptors. In our study, pretreatment of neutrophils with anti-CD82 Abs had no apparent effect on resting calcium homeostasis, whereas POAEC-induced calcium rises were significantly reduced. This ap- parent discrepancy may be due to different source of Abs and different

FIGURE 5. Expression of CD82 in undifferentiated and differentiated cell lines as measured by quantitative RT-PCR (A,upper panel), by Western blots (A,lower panel), and by confocal laser scanning microscopy (B). Data shown are representative of at least three independent experiments. Data in quantitative RT-PCR are expressed as means6SEM. Colocalization of CD82 (green) and LFA-1a(red) is shown in (C). All confocal micrographs were performed on live cells labeled on ice and staining performed as indicated inMaterials and Methods. Original magnification340.

FIGURE 6. Inhibition of POAEC-in- duced calcium transients in human naive neutrophils by Abs to CD82. (A) calcium maps in control (isoform-treated) cells and (B) in anti-CD82–treated cells. Calcium changes are color-coded such that high calcium is seen as warm colors. (C) Fluo- rescence intensities were transformed into absolute calcium levels in control (black) and anti-CD82–treated (red) human naive neutrophils. Data are representative of at least five independent experiments. Calcium values are expressed as means6SEM (n= 11 and 13 cells for control and anti-CD82–

treated cells, respectively). (D) Inhibition of POAEC-induced ROM production by anti- CD82 Abs (red) as compared with control (blue) as measured by LDCL. Staining was performed as indicated in Materials and Methods. Original magnification340.

(10)

cell types. Interestingly, human naive neutrophils lost their adhesive capabilities following anti-CD82 Ab treatment (data not shown).

The question as to the CD82 counterligand on POAECs is yet to be determined. A clue may be provided by the presence of anti–non-Gal xenoreactive Abs in human and nonhuman primates. It is estimated that 10% of total anti-porcine Abs are targeted against the non-Gal moieties (30). Interestingly, GalT KO kidneys have been demon- strated to elicit xenoantibody response in nonhuman primates (29, 58). Although no clear targets were identified, carbohydrate Ags on the xenogeneic endothelium might be a plausible one. The presence of anti-Hanganutziu–Deicher xenoantibodies suggests a role for sialic acid Ags. Sialic acid normally exists as N-ace- tyleneuromic acid or N-glucolylneuramic acid. The latter is not expressed in humans (30). The association of CD82 and other tetraspanins with gangliosides is known to modulate cellular signaling (59, 60). Whether CD82 on human neutrophils in- teracts directly with the sialic acid–containing structures on the porcine endothelium remains to be answered. Additionally, the significance of this “newly identified receptor” to the overall re- jection process must be considered, especially in the presence of anti-porcine Abs, complement, and platelets. Tetraspanins are membrane organizers and are associated with a variety of mole- cules, including complement receptor 2 (51), glycosynapse (59), and gangliosides, which are avid organizers of CD82-enriched microdomains (60).

Although the data presented in this study assign a new role for CD82 in xenorecognition and may therefore suggest that its downregulation is desirable for xenograft survival, such down- regulation has been associated with increased metastasis in a number of human cancers, including colorectal cancers (61), squamous cell carcinoma (62), prostate cancer (38), breast cancer (63), and hepatocellular carcinoma (64, 65). The question as to how CD82, a tumor suppressor protein, mediates xenorecognition is yet to be addressed. Tumor metastasis suppression by CD82 is thought to occur by stabilizing E-cadherin/b-catenin complex formation (66), upregulation of Sprouty2 (67), maturation of

b1

- integrin (68), and direct interactions with the Duffy Ag/receptor for chemokine–expressing endothelial cells (69). Whether such mechanisms could be exploited for increased survival of a xeno- graft remains to be addressed. Of particular interest is the Duffy Ag/receptor for chemokine, which was first discovered as a blood group Ag, and then was found to be a receptor for malaria para- sites on erythrocytes and a chemokine receptor. It has been in- volved in allogeneic rejection of lung and renal transplants (70, 71). This CD8-binding and sialic acid–containing glycoprotein has been associated with neutrophil activation through NAP-2 and NAP-3–dependent mechanisms (72). It is therefore tempting to speculate that CD82–Duffy Ag/receptor for chemokine interactions may also occur during the xenogeneic encounter. This remains to be tested. In any case, the widespread expression of CD82 in the recipient innate immune cells (neutrophils, macrophages, NK cells, and others) may constitute a more formidable barrier for long-term survival of a transplanted xenograft.

Acknowledgments

We express gratitude to Hakem Al-Enazi for his efforts to get supplies and reagents on time. We thank Drs. M. Al-Joufan, M. Akhtar, and R. Sougrat for valuable discussions. We are also indebted to the research center ad- ministration for continuous support.

Disclosures

The authors have no financial conflicts of interest.

References

1. Stehlik, J., L. B. Edwards, A. Y. Kucheryavaya, C. Benden, J. D. Christie, F. Dobbels, R. Kirk, A. O. Rahmel, and M. I. Hertz. 2011. The registry of the International Society for Heart and Lung Transplantation: Twenty-Eighth Adult Heart Transplant Report: 2011.J. Heart Lung Transplant.30: 1078–1094.

2. Cooper, D. K. 1992. Depletion of natural antibodies in non-human primates:

a step towards successful discordant xenografting in humans.Clin. Transplant.6:

178–183.

3. Saadi, S., and J. L. Platt. 1995. Transient perturbation of endothelial integrity induced by natural antibodies and complement.J. Exp. Med.181: 21–31.

4. Good, A. H., D. K. Cooper, A. J. Malcolm, R. M. Ippolito, E. Koren, F. A. Neethling, Y. Ye, N. Zuhdi, and L. R. Lamontagne. 1992. Identification of carbohydrate structures that bind human antiporcine antibodies: implications for discordant xenografting in humans.Transplant. Proc.24: 559–562.

5. Cooper, D. K. 2010. Outwitting evolution.Xenotransplantation17: 171–180.

6. Galili, U., E. A. Rachmilewitz, A. Peleg, and I. Flechner. 1984. A unique natural human IgG antibody with anti-alpha-galactosyl specificity.J. Exp. Med.160:

1519–1531.

7. Byrne, G. W., and C. G. A. McGregor. 2012. Cardiac xenotransplantation:

progress and challenges.Curr. Opin. Organ Transplant.17: 148–154.

8. Katopodis, A. G., R. G. Warner, R. O. Duthaler, M. B. Streiff, A. Bruelisauer, O. Kretz, B. Dorobek, E. Persohn, H. Andres, A. Schweitzer, et al. 2002. Re- moval of anti-Gala1,3Gal xenoantibodies with an injectable polymer.J. Clin.

Invest.110: 1869–1877.

9. Maruyama, S., E. Cantu, III, C. DeMartino, C. Y. Wang, J. Chen, F. Al-Mohanna, S. M. Nakeeb, V. D’Agati, B. Pernis, U. Galili, et al. 1999. Interaction of baboon anti-a-galactosyl antibody with pig tissues.Am. J. Pathol.155: 1635–1649.

10. Murakami, H., H. Nagashima, Y. Takahagi, S. Miyagawa, T. Fujimura, K. Toyomura, R. Nakai, M. Yamada, T. Kurihara, T. Shigehisa, et al. 2002.

Transgenic pigs expressing human decay-accelerating factor regulated by por- cine MCP gene promoter.Mol. Reprod. Dev.61: 302–311.

11. Nottle, M. B., L. F. Beebe, S. J. Harrison, S. M. McIlfatrick, R. J. Ashman, P. J. O’Connell, E. J. Salvaris, N. Fisicaro, S. Pommey, P. J. Cowan, and A. J. d’Apice. 2007. Production of homozygous a-1,3-galactosyltransferase knockout pigs by breeding and somatic cell nuclear transfer.Xenotransplantation 14: 339–344.

12. Zhu, M., W. Zhang, F. Liu, L. Wang, B. Liu, D. Chen, X. H. Zhu, W. J. Zhang, T. E. Ichim, Z. S. Chen, et al. 2008. Resistance to anti-xenogeneic response by combininga-Gal silencing with HO-1 upregulation.Transpl. Immunol.19: 202–

208.

13. Kuwaki, K. Y., Y. L. Tseng, F. J. Dor, A. Shimizu, S. L. Houser, T. M. Sanderson, C. J. Lancos, D. D. Prabharasuth, J. Cheng, K. Moran, et al. 2005. Heart transplantation in baboons usinga1,3-galactosyltransferase gene-knockout pigs as donors: initial experience.Nat. Med.11: 29–31.

14. Yamada, K., K. Yazawa, A. Shimizu, T. Iwanaga, Y. Hisashi, M. Nuhn, P. O’Malley, S. Nobori, P. A. Vagefi, C. Patience, et al. 2005. Marked prolon- gation of porcine renal xenograft survival in baboons through the use ofa1,3- galactosyltransferase gene-knockout donors and the cotransplantation of vascu- larized thymic tissue.Nat. Med.11: 32–34.

15. Ekser, B., M. Ezzelarab, H. Hara, D. J. van der Windt, M. Wijkstrom, R. Bottino, M. Trucco, and D. K. Cooper. 2012. Clinical xenotransplantation: the next medical revolution?Lancet379: 672–683.

16. al-Mohanna, F., K. Collison, R. Parhar, A. Kwaasi, B. Meyer, S. Saleh, S. Allen, S. al-Sedairy, D. Stern, and M. Yacoub. 1997. Activation of naive xenogeneic but not allogeneic endothelial cells by human naive neutrophils: a potential occult barrier to xenotransplantation.Am. J. Pathol.151: 111–120.

17. Cardozo, L. A., D. B. Rouw, L. R. Ambrose, M. Midulla, O. Florey, D. O. Haskard, and A. N. Warrens. 2004. The neutrophil: the unnoticed threat in xenotransplantation?Transplantation78: 1721–1728.

18. Al-Mohanna, F., S. Saleh, R. S. Parhar, K. Khabar, and K. Collison. 2005.

Human neutrophil gene expression profiling following xenogeneic encounter with porcine aortic endothelial cells: the occult role of neutrophils in xenograft rejection revealed.J. Leukoc. Biol.78: 51–61.

19. Sheikh, S., R. S. Parhar, A. Kwaasi, K. Collison, M. Yacoub, D. Stern, and F. Al- Mohanna. 2000.a-Gal-independent dual recognition and activation of xenoge- neic endothelial cells and human naı¨ve natural killer cells.Transplantation70:

917–928.

20. Baumann, B. C., M. K. Schneider, B. G. Lilienfeld, M. A. Antsiferova, D. M. Rhyner, R. J. Hawley, and J. D. Seebach. 2005. Endothelial cells derived from pigs lacking Gala(1,3)Gal: no reduction of human leukocyte adhesion and natural killer cell cytotoxicity.Transplantation79: 1067–1072.

21. Phelps, C. J., C. Koike, T. D. Vaught, J. Boone, K. D. Wells, S. H. Chen, S. Ball, S. M. Specht, I. A. Polejaeva, J. A. Monahan, et al. 2003. Production ofa1,3- galactosyltransferase-deficient pigs.Science299: 411–414.

22. Seebach, J. D., M. K. Schneider, C. A. Comrack, A. LeGuern, S. A. Kolb, P. A. Knolle, S. Germana, H. DerSimonian, C. LeGuern, and D. H. Sachs. 2001.

Immortalized bone-marrow derived pig endothelial cells.Xenotransplantation8:

48–61.

23. Ishibashi, K., S. Okazaki, and M. Hiramatsu. 2006. Simultaneous measurement of superoxide generation and intracellular Ca2+concentration reveals the effect of extracellular Ca2+on rapid and transient contents of superoxide generation in differentiated THP-1 cells.Biochem. Biophys. Res. Commun.344: 571–580.

24. Al-Mohanna, F. A., and M. B. Hallett. 1988. The use of fura-2 to determine the relationship between cytoplasmic free Ca2+and oxidase activation in rat neu- trophils.Cell Calcium9: 17–26.

2804 ROLE OF CD82 IN INTERCELLULAR XENOGENEIC MOLECULAR DIALOGUE

(11)

25. Lash, A. E. C. M., C. M. Tolstoshev, L. Wagner, G. D. Schuler, R. L. Strausberg, G. J. Riggins, and S. F. Altschul. 2000. SAGEmap: a public gene expression resource.Genome Res.10: 1051–1060.

26. Collins, S. J., F. W. Ruscetti, R. E. Gallagher, and R. C. Gallo. 1979. Normal functional characteristics of cultured human promyelocytic leukemia cells (HL- 60) after induction of differentiation by dimethylsulfoxide.J. Exp. Med.149:

969–974.

27. Thompson, B. Y., G. Sivam, B. E. Britigan, G. M. Rosen, and M. S. Cohen. 1988.

Oxygen metabolism of the HL-60 cell line: comparison of the effects of mon- ocytoid and neutrophilic differentiation.J. Leukoc. Biol.43: 140–147.

28. Momose, H., H. Kurosu, N. Tsujimoto, K. Kontani, K. Tsujita, H. Nishina, and T. Katada. 2003. Dual phosphorylation of phosphoinositide 3-kinase adaptor Grb2-associated binder 2 is responsible for superoxide formation synergistically stimulated by Fcgand formyl-methionyl-leucyl-phenylalanine receptors in dif- ferentiated THP-1 cells.J. Immunol.171: 4227–4234.

29. Chen, G., H. Qian, T. Starzl, H. Sun, B. Garcia, X. Wang, Y. Wise, Y. Liu, Y. Xiang, L. Copeman, et al. 2005. Acute rejection is associated with antibodies to non-Gal antigens in baboons using Gal-knockout pig kidneys.Nat. Med.11:

1295–1298.

30. Saethre, M., B. C. Baumann, M. Fung, J. D. Seebach, and T. E. Mollnes. 2007.

Characterization of natural human anti-non-Gal antibodies and their effect on activation of porcine Gal-deficient endothelial cells.Transplantation84: 244–

250.

31. Lebel-Binay, S., C. Lagaudrie`re, D. Fradelizi, and H. Conjeaud. 1995. CD82, tetra-span-transmembrane protein, is a regulated transducing molecule on U937 monocytic cell line.J. Leukoc. Biol.57: 956–963.

32. Peng, W. M., C. F. Yu, W. Kolanus, A. Mazzocca, T. Bieber, S. Kraft, and N. Novak. 2011. Tetraspanins CD9 and CD81 are molecular partners of trimeric FcεRI on human antigen-presenting cells.Allergy66: 605–611.

33. Mela, A., and J. E. Goldman. 2009. The tetraspanin KAI1/CD82 is expressed by late-lineage oligodendrocyte precursors and may function to restrict precursor migration and promote oligodendrocyte differentiation and myelination. J.

Neurosci.29: 11172–11181.

34. Xu, C., Y. H. Zhang, M. Thangavel, M. M. Richardson, L. Liu, B. Zhou, Y. Zheng, R. S. Ostrom, and X. A. Zhang. 2009. CD82 endocytosis and cholesterol-dependent reorganization of tetraspanin webs and lipid rafts.FASEB J.23: 3273–3288.

35. Lee, H. A., I. Park, H. J. Byun, D. Jeoung, Y. M. Kim, and H. Lee. 2011. Me- tastasis suppressor KAI1/CD82 attenuates the matrix adhesion of human prostate cancer cells by suppressing fibronectin expression andb1 integrin activation.

Cell. Physiol. Biochem.27: 575–586.

36. Voss, M. A., N. Gordon, S. Maloney, R. Ganesan, L. Ludeman, K. McCarthy, R. Gornall, G. Schaller, W. Wei, F. Berditchevski, and S. Sundar. 2011. Tetra- spanin CD151 is a novel prognostic marker in poor outcome endometrial cancer.

Br. J. Cancer104: 1611–1618.

37. Berditchevski, F., E. Gilbert, M. R. Griffiths, S. Fitter, L. Ashman, and S. J. Jenner. 2001. Analysis of the CD151-a3b1integrin and CD151-tetraspanin interactions by mutagenesis.J. Biol. Chem.276: 41165–41174.

38. Liu, W., M. Iiizumi-Gairani, H. Okuda, A. Kobayashi, M. Watabe, S. K. Pai, P. R. Pandey, F. Xing, K. Fukuda, V. Modur, et al. 2011. KAI1 gene is engaged in NDRG1 gene-mediated metastasis suppression through the ATF3-NFkB com- plex in human prostate cancer.J. Biol. Chem.286: 18949–18959.

39. Huang, S., S. Yuan, M. Dong, J. Su, C. Yu, Y. Shen, X. Xie, Y. Yu, X. Yu, S. Chen, et al. 2005. The phylogenetic analysis of tetraspanins projects the evolution of cell-cell interactions from unicellular to multicellular organisms.

Genomics86: 674–684.

40. Luo, R. F., S. Zhao, R. Tibshirani, J. H. Myklebust, M. Sanyal, R. Fernandez, D. Gratzinger, R. J. Marinelli, Z. S. Lu, A. Wong, et al. 2010. CD81 protein is expressed at high levels in normal germinal center B cells and in subtypes of human lymphomas.Hum. Pathol.41: 271–280.

41. Chirumbolo, S., A. Conforti, R. Ortolani, A. Vella, M. Marzotto, and P. Bellavite.

2010. Stimulus-specific regulation of CD63 and CD203c membrane expression in human basophils by the flavonoid quercetin.Int. Immunopharmacol.10: 183–

192.

42. Novitskaya, V., H. Romanska, M. Dawoud, J. L. Jones, and F. Berditchevski.

2010. Tetraspanin CD151 regulates growth of mammary epithelial cells in three- dimensional extracellular matrix: implication for mammary ductal carcinoma in situ.Cancer Res.70: 4698–4708.

43. Protty, M. B., N. A. Watkins, D. Colombo, S. G. Thomas, V. L. Heath, J. M. Herbert, R. Bicknell, Y. A. Senis, L. K. Ashman, F. Berditchevski, et al. 2009.

Identification of Tspan9 as a novel platelet tetraspanin and the collagen receptor GPVI as a component of tetraspanin microdomains.Biochem. J.417: 391–400.

44. Nagao, K., and K. Oka. 2011. HIF-2 directly activates CD82 gene expression in endothelial cells.Biochem. Biophys. Res. Commun.407: 260–265.

45. Cao, J., P. Zhao, L. J. Zhao, S. M. Wu, S. Y. Zhu, and Z. T. Qi. 2003. Identi- fication and expression of human CD81 gene on murine NIH/3T3 cell mem- brane.J. Microbiol. Methods54: 81–85.

46. Wang, H., W. Zhang, J. Zhao, L. Zhang, M. Liu, G. Yan, J. Yao, H. Yu, and P. Yang. 2012.N-glycosylation pattern of recombinant human CD82 (KAI1), a tumor-associated membrane protein.J. Proteomics75: 1375–1385.

47. Lazo, P. 2009. Roles of tetraspanin proteins in cell and tumor biology.Atlas Genet. Cytogenet. Oncol. Haematol.13: 1151–1172.

48. Delaguillaumie, A., J. Harriague, S. Kohanna, G. Bismuth, E. Rubinstein, M. Seigneuret, and H. Conjeaud. 2004. Tetraspanin CD82 controls the associ- ation of cholesterol-dependent microdomains with the actin cytoskeleton in T lymphocytes: relevance to co-stimulation.J. Cell Sci.117: 5269–5282.

49. Berditchevski, F. 2001. Complexes of tetraspanins with integrins: more than meets the eye.J. Cell Sci.114: 4143–4151.

50. Wang, X. Q., Q. Yan, P. Sun, J. W. Liu, L. Go, S. M. McDaniel, and A. S. Paller.

2007. Suppression of epidermal growth factor receptor signaling by protein ki- nase C-aactivation requires CD82, caveolin-1, and ganglioside.Cancer Res.67:

9986–9995.

51. Cherukuri, A., T. Shoham, H. W. Sohn, S. Levy, S. Brooks, R. Carter, and S. K. Pierce. 2004. The tetraspanin CD81 is necessary for partitioning of coli- gated CD19/CD21-B cell antigen receptor complexes into signaling-active lipid rafts.J. Immunol.172: 370–380.

52. Charrin, S., S. Manie´, C. Thiele, M. Billard, D. Gerlier, C. Boucheix, and E. Rubinstein. 2003. A physical and functional link between cholesterol and tetraspanins.Eur. J. Immunol.33: 2479–2489.

53. Mittelbrunn, M., M. Ya´n˜ez-Mo´, D. Sancho, A. Ursa, and F. Sa´nchez-Madrid. 2002.

Cutting edge: dynamic redistribution of tetraspanin CD81 at the central zone of the immune synapse in both T lymphocytes and APC.J. Immunol.169: 6691–6695.

54. Davis, D. M., and M. L. Dustin. 2004. What is the importance of the immu- nological synapse?Trends Immunol.25: 323–327.

55. Sigalov, A. E. 2008. Multichain immune recognition receptor signaling from spatiotemporal organization to human disease.Adv. Exp. Med. Biol.640: ix–xi.

56. Worthington, R. E., R. C. Carroll, and C. Boucheix. 1990. Platelet activation by CD9 monoclonal antibodies is mediated by the Fc gamma II receptor.Br. J.

Haematol.74: 216–222.

57. Olweus, J., F. Lund-Johansen, and V. Horejsi. 1993. CD53, a protein with four membrane-spanning domains, mediates signal transduction in human monocytes and B cells.J. Immunol.151: 707–716.

58. Harnden, I., K. Kiernan, and M. Kearns-Jonker. 2010. The anti-nonGal xen- oantibody response toa1,3-galactosyltransferase gene knockout pig xenografts.

Curr. Opin. Organ Transplant.15: 207–211.

59. Todeschini, A. R., J. N. Dos Santos, K. Handa, and S. I. Hakomori. 2007.

Ganglioside GM2-tetraspanin CD82 complex inhibits met and its cross-talk with integrins, providing a basis for control of cell motility through glycosynapse.J.

Biol. Chem.282: 8123–8133.

60. Odintsova, E., T. D. Butters, E. Monti, H. Sprong, G. van Meer, and F. Berditchevski. 2006. Gangliosides play an important role in the organization of CD82-enriched microdomains.Biochem. J.400: 315–325.

61. Maurer, C. A., H. U. Graber, H. Friess, B. Beyermann, D. Willi, P. Netzer, A. Zimmermann, and M. W. Bu¨chler. 1999. Reduced expression of the metas- tasis suppressor gene KAI1 in advanced colon cancer and its metastases.Surgery 126: 869–880.

62. Kawasaki, G., I. Yoshitomi, S. Yanamoto, S. Yamada, A. Mizuno, and M. Umeda. 2011. Expression of thymidylate synthase and dihydropyrimidine dehydrogenase in primary oral squamous cell carcinoma and corresponding metastases in cervical lymph nodes: association with the metastasis suppressor CD82.Anticancer Res.31: 3521–3526.

63. Malik, F. A., A. J. Sanders, M. A. Kayani, and W. G. Jiang. 2009. Effect of expressional alteration of KAI1 on breast cancer cell growth, adhesion, migra- tion and invasion.Cancer Genomics Proteomics6: 205–213.

64. Yang, J. M., Z. H. Peng, S. H. Si, W. W. Liu, Y. H. Luo, and Z. Y. Ye. 2008.

KAI1 gene suppresses invasion and metastasis of hepatocellular carcinoma MHCC97-H cells in vitro and in animal models.Liver Int.28: 132–139.

65. Tsai, Y. C., and A. M. Weissman. 2011. Dissecting the diverse functions of the metastasis suppressor CD82/KAI1.FEBS Lett.585: 3166–3173.

66. Abe, M., T. Sugiura, M. Takahashi, K. Ishii, M. Shimoda, and K. Shirasuna.

2008. A novel function of CD82/KAI-1 on E-cadherin-mediated homophilic cellular adhesion of cancer cells.Cancer Lett.266: 163–170.

67. Mu, Z., H. Wang, J. Zhang, Q. Li, L. Wang, and X. Guo. 2008. KAI1/CD82 suppresses hepatocyte growth factor-induced migration of hepatoma cells via upregulation of Sprouty2.Sci. China C Life Sci.51: 648–654.

68. Jee, B. K., J. Y. Lee, Y. Lim, K. H. Lee, and Y. H. Jo. 2007. Effect of KAI1/CD82 on theb1integrin maturation in highly migratory carcinoma cells.Biochem.

Biophys. Res. Commun.359: 703–708.

69. Bandyopadhyay, S., R. Zhan, A. Chaudhuri, M. Watabe, S. K. Pai, S. Hirota, S. Hosobe, T. Tsukada, K. Miura, Y. Takano, et al. 2006. Interaction of KAI1 on tumor cells with DARC on vascular endothelium leads to metastasis suppression.

Nat. Med.12: 933–938.

70. Geleff, S., D. Draganovici, P. Jaksch, and S. Segerer. 2010. The role of che- mokine receptors in acute lung allograft rejection.Eur. Respir. J.35: 167–175.

71. Watorek, E., M. Boratynska, A. Hałon, and M. Klinger. 2008. Anti-Fya anti- bodies as the cause of an unfortunate post-transplant course in renal transplant recipient.Ann. Transplant.13: 48–52.

72. Horne, K., and I. J. Woolley. 2009. Shedding light on DARC: the role of the Duffy antigen/receptor for chemokines in inflammation, infection and malig- nancy.Inflamm. Res.58: 431–435.

Références

Documents relatifs

To prevent differentiation, we now aim to overexpress transcription factors involved in sustaining pluripotency in mouse and human ES cells, namely Oct4, Sox2

Analysis of an independent panel of genes selected by the International Stem Cell Initiative 37 showed that the TL2i cells and ground-state mESCs formed a distinct cluster

Deregulation of intracellular lipid metabolism has been extensively studied in foamy macrophages in the con- text of atherosclerosis ( 4 ). However, much less is known regarding

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des

Classical Polarizable Force Field To Study Hydrated Charged Clays and Zeolites.. Stéphane Tesson, Wilfried Louisfrema, Mathieu Salanne, Anne Boutin, Eric Ferrage, Benjamin

Gulati, R. G.: Anomalous carbon iso- tope fractionation between atmospheric CO 2 and dissolved inor- ganic carbon induced by intense photosynthesis, Geochim. and Meijer,

One Ontario board phased out a school by extending boundaries of adjacent schools until aIl the students had been absorbed into other schools.. As the costs of