Intercellular adhesion molecule (ICAM)-1 is a single chain glycoprotein that is a member of the immunoglobulin superfamily and in humans acts as a ligand for the β2 integrins leukocyte function-associated antigen (LFA)-1 (β2αL; CD18/CD11a) and Mac-1 (β2αM; CD18/CD11b) as well as fibrinogen (1–7). Since ICAM-1 was first characterized, it has been identified in several species other than human, including dog, rat, mouse, and cow (8–12). There is now a large literature demonstrating the importance of ICAM-1 for leukocyte interactions with other cells that facilitate leukocyte adhesion, migration, activation, and effector function (13).
Adhesion of leukocytes to endothelial cells (EC) is the first step in their migration into inflamed tissues (13) and is also of paramount importance during the immunological rejection of allografts (14–19). The binding of β2 integrins to ICAM-1 is thought to be important for establishing stable adhesions of leukocytes that become activated during selectin-mediated rolling on the EC surface and may also be involved in the transmigration of leukocytes through endothelium into tissues (13). In the case of monocytes, lymphocytes, and eosinophils, α4 integrins also participate in these events through binding vascular cell adhesion molecule (VCAM)-1, another single chain glycoprotein that has structural similarity to ICAM-1 (20–22). Expression of both ICAM-1 and VCAM-1 by human EC is regulated by cytokines (23).
The shortage of human organs has led to increasing interest in the use of porcine organs for clinical xenotransplanation (24, 25), and there is therefore a need to define the mechanisms by which human leukocytes will interact with porcine endothelium. Previous reports have characterized VCAM-1 on porcine EC, both in vitro and in vivo (26, 27). In contrast, evidence for the existence of ICAM-1 in the pig has remained incomplete. Thus, although monoclonal antibodies (mAb) to human LFA-1 have been shown to inhibit the interactions of human lymphocytes with porcine cells (28–31), the precise porcine ligands involved have not yet been determined. Furthermore, although an anti-human ICAM-1 mAb (clone 11C8) has been reported to react with interleukin (IL)-1-activated porcine chondrocytes (32), cross-reactivity of this or other anti-ICAM-1 mAb with a porcine homologue has been difficult to establish in our hands. We therefore undertook to isolate porcine ICAM-1 cDNA and generate anti-porcine ICAM-1 mAb with which to characterize expression of this molecule by porcine EC.
Recombinant human tumor necrosis factor (rhTNF)-α was a kind gift from Dr. Gary Jesmok (Bayer Corporation, West Haven CT); rhIL-1α was a gift from Dr. Jean-Jacques Mermot (Glaxo Institute of Molecular Biology, Geneva, Switzerland); rhIL-4 was purchased from Genzyme (Boston, MA); rhIL-2 was purchased from Boehringer Mannheim (Lewes, East Sussex, UK); recombinant human interferon (rhIFN)-γ was from Biogen (Cambridge, MA); recombinant porcine (rp) IFN-γ was purchased from Innogenetics (Ghent, Belgium); and rpIL-4 was from Bioscource (Appligene-Oncor Lifescreen, Watford, UK). Lipopolysaccharide (Escherichia coli serotype 0111:B4) was from Sigma (Poole, Dorset, UK). The cytokine/lipopolysaccharide-stimulated porcine aortic endothelial cell (PAEC) cDNA library in pKS1 was a kind gift from Dr. Martyn Robinson (Celltech, Berkshire, UK), as were the full-length cDNAs of porcine E-selectin (3.3 kb) and VCAM-1 (3 kb). The full-length human (2.9 kb) ICAM-1 cDNA in pCDM8 was a kind gift from Dr. David Simmons (SmithKline Pharmaceuticals, Harlow, Essex, UK). The rat GAPDH cDNA (1.6 kb), which cross-reacts with the porcine sequence, was a kind gift from Dr. R de Martin (Vienna International Research Cooperation Centre, Vienna, Austria).
The nonbinding control mAb G155–228 (IgM; anti-TNP-keyhole limpet hemocyanin) was purchased from PharMingen (San Diego, CA). The nonbinding control antibody MOPC21 (IgG1), mAb 6.5E (IgG1, anti-human CD18/β2 integrin), mAb HP1/2 (IgG1, anti-human CD49d/α4 integrin), and mAb 10.2C7 (IgG1, anti-porcine VCAM-1) (27) were kind gifts from Dr. Martyn Robinson (Celltech Ltd., Slough, UK). mAb 15.2 (IgG1, anti-human ICAM-1) was a gift from Dr. Nancy Hogg, Imperial Cancer Research Fund, London, UK. mAb LC-14 (anti-porcine CD31) was a kind gift from Dr. Peter Kilshaw (The Babraham Institute, Cambridge, UK). mAb MA251 (anti-human CD25) was purchased from Serotec (Oxford, UK). mAb K231 (anti-porcine CD25.3B2) was a kind gift from Dr. Tony Whyte (The Babraham Institute, Cambridge, UK). Mouse anti-human Ig (Fab specific, mAb GG6) was purchased from Sigma.
Polymerase chain reaction (PCR).
PCR were performed in a 50- to 100-μl volume using 10 pg to 100 ng of plasmid or 100 ng to 1 μg of cDNA library DNA as template. The reaction conditions were as follows: 10 pmol of each oligonucleotide, 200 μM dNTPs, and 2.5–5 U of Taq DNA polymerase (Amplitaq Gold; Perkin Elmer, Warrington, Cheshire, UK) in the manufacturer’s reaction buffer. A total of 30–35 cycles were performed using a melting temperature of 94°C for 1 min, annealing temperature of 50–62°C (optimized experimentally) for 30–60 sec, and extension at 72°C for 1–2 min (depending on product length). Samples were analyzed by agarose gel electrophoresis.
The reverse transcription (RT) reaction utilized 1 μg of mRNA from activated PAEC. The mRNA was heated in 10 μl of RNase free water to 70°C for 5 min and then snap-chilled on ice. This was then reacted with 10 mM dNTPs, 2 μl of 0.1 M dithiothreitol, 1× first-strand buffer, 0.5 μg of oligo(dT) 12–18, 200 U of Superscript (Gibco, Life Technologies, Paisley, Scotland), and 2 U of Rnase Guard (Pharmacia, Uppsala, Sweden) at 42°C for 55 min. The reaction was then inactivated by heating at 70°C for 15 min. The RT reaction was then used as a cDNA template in a 100-μl PCR reaction using 5 U of Taq DNA polymerase (Gibco) and 3 mM MgCl2 and cycled as follows: 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min for 35 cycles.
Transfection of COS-7 cells.
COS-7 cells (4×106) were cultured until approximately 60% confluent, harvested, and electroporated (290 V, 1500 μF) in RPMI 1640, antibiotics (100 IU/ml penicillin, 100 μg/ml streptomycin, 100 μg/ml gentamycin, 2 mM l-glutamine, and 10% fetal calf serum (FCS) (all from Gibco) with 30–50 μg of cDNA. Cells were then plated onto 9-cm tissue culture dishes (Nunc, Rosehilde, Denmark) and cultured for 24 hr, whereupon the medium was replaced and the electroporated cells were cultured for a further 24 hr. For some experiments, cells were harvested 24 hr after transfection, plated at 5×104 cells onto 35-mm tissue culture dishes (Nunc), and used in adhesion assays 48 hr later.
Preparation of phytohemagglutinin (PHA)-stimulated lymphoblasts.
Human or porcine mononuclear cells (MNC) were isolated from peripheral blood by density gradient centrifugation on Ficoll-Hypaque as previously described (33). After culture for 24 hr, MNC were stimulated for 48 hr at 37°C with 2 μg/ml Phaseolus Vulgaris PHA (Sigma). Cells were assessed for activation on the basis of blast morphology and by cell surface antigen staining with anti-CD25 mAb. Cell preparations were harvested by pipetting, washed twice with Hanks’ balanced salt solution (HBSS) without calcium and magnesium to disperse large cell aggregates, and resuspended in adjuvant before use for immunization. To prepare lymphoblasts for adhesion assays, PHA-stimulated MNC were cultured for a further 14 days in growth medium containing 10 U/ml hrIL-2.
Isolation and culture of EC.
PAEC and human umbilical vein EC (HUVEC) were isolated as previously described (26, 34). EC were cultured in 25-cm2 flasks (Costar, High Wycombe, UK) precoated with 1% gelatin (Sigma) at 37°C, 5% CO2 in growth medium consisting of RPMI 1640 supplemented with 20% heat-inactivated FCS (56°C for 30 min), antibiotics, 15 U/ml sodium heparin, 2 mM l-glutamine (all from Gibco), and 10 μg/ml endothelial cell growth factor (Sigma). Once confluent, cells were washed twice with HBSS without calcium and magnesium and detached with trypsin/EDTA (ICN Biomedicals Inc, Costa Mesa, CA) or cell dissociation nonenzymatic mixture (Sigma). EC cultures were maintained for up to five passages.
Plating and stimulation of EC for surface antigen expression assays.
EC were detached from tissue culture flasks with 0.125% trypsin-EDTA, resuspended in growth medium, and cultured for at least 24 hr in 96-well microtiter plates (3–4×104 cells/well), 24-well plates (2–3×105 cells/well), or 35-mm petri dishes (7×105 per dish) to achieve confluent monolayers. Before stimulation, the medium was replaced with culture medium containing no growth supplements. Stimulation of EC was performed by adding the appropriate volume of the stimulant at 10 times the desired final concentration to minimize disturbance of the cells. After stimulation with TNF-α (10 ng/ml for 16 hr), EC in 96-well plates to be used for screening hybridoma supernatants were fixed with 2% paraformaldehyde-lysine-periodate (35) and stored at 4°C for up to 3 weeks.
Generation of mAb.
Porcine PHA-stimulated lymphoblasts were washed three times in HBSS and then reconstituted 1:1 in MPL-TDM adjuvant (Sigma) in normal saline according to the manufacturer’s protocol. A Balb/c mouse was immunized three times i.p. (4×106 cells in 100 μl/mouse), with 2 weeks between the first injection and 5 weeks between the second and third. Three days after the third immunization, splenocytes were fused with the NSO nonsecreting myeloma cell line, as previously described (34). Hybridoma supernatants were screened by ELISA for reactivity with TNF-α-activated PAEC (34). Positive hybridoma cultures were cloned twice by limiting dilution before studying mAb reactivity further. The mAb isotype was determined using a mouse mAb isotyping kit (Serotec) according to the manufacturer’s protocol.
Northern analysis and hybridization.
Northern analysis was performed as described by Sambrook et al. (36) with apparatus and materials treated with diethylpyrocarbonate (Sigma). RNA was electrophoresed through a 1% agarose gel containing a final 6.6% (v/v) formaldehyde in 1× MOPS buffer, pH 7 (5× MOPS buffer: 200 mM 3-N-morpholino propane-sulphonic acid, 50 mM sodium acetate, and 5 mM EDTA, pH 8; Sigma). The RNA (10 μg per lane) was dissolved in 50% redistilled formamide (Gibco), 6.6% formaldehyde, and 0.5× MOPS buffer and heated to 65°C for 10 min. Sample loading buffer (50% glycerol, 0.25% xylene cyanol, 0.25% bromophenol blue, and 1 mM EDTA) containing 1 μg of ethidium bromide was added to each sample, and electrophoresis was carried out at 60–80 V for 1–2 hr in 1× MOPS buffer. The RNA was visualized under ultraviolet light and blotted onto a Duralon membrane (Stratagene, La Jolla, CA) by capillary blotting and immobilized by ultraviolet cross-linking. Probes were labeled with [32P]dCTP using the Quickprime kit (Pharmacia) according to the manufacturer’s instructions. Hybridization with a radiolabeled probe was carried out at 42°C overnight. The blot was washed and autoradiographed using Hyperfilm (Nycomed Pharmacia, Amersham, UK).
EC were harvested and resuspended in growth medium at 4°C. After repeated pipetting to ensure single-cell suspensions, 105-106 cells were plated in V-bottom 96-well plates (Costar, High Wycombe, UK) and centrifuged at 200×g for 5 min. The supernatants were carefully flicked off and the pellets resuspended with 100 μl of saturating amounts of the appropriate primary mAb, either hybridoma culture supernatant or 20 μg/ml purified antibody, at 4°C in HBSS, 2.5% FCS, and 0.1% azide for 30 min. Cells were washed, centrifuged for 10 min at 1400 rpm at 4°C, and then resuspended in a fluorescein-conjugated rabbit anti-mouse IgG (Dako, High Wycombe, UK) 1:50 in HBSS/2.5% FCS/0.1% azide at 4°C for 30 min. Cells were then centrifuged and washed twice as above and fixed with 1% paraformaldehyde v/v in phosphate-buffered saline (PBS) and analyzed using a Epics XL flow cytometer (Coulter Electronics Ltd., Luton, UK). Dead cells were excluded by setting appropriate forward and 90o side scatter gates. Positive cells were determined by setting a threshold with reference to the relevant negative control stained with isotype-matched irrelevant antibody. In some experiments, specific mAb staining was quantified by normalizing the mean fluorescent staining intensity obtained with test antibody to that obtained with an appropriate isotype-matched control antibody, thereby establishing a relative fluorescence staining intensity (RFI, where an RFI of 1.00=no expression).
Lymphoblast adhesion to COS-7 cell transfectants.
COS-7 cells were transfected with cDNA and cultured for 1 day before harvesting and plating on 35-mm dishes. Before adhesion assays, COS-7 cell transfectants were fixed with 2% formaldehyde in PBS for 1 hr at room temperature. Dishes were then washed six times for 10 min with PBS. Lymphoblasts were added to COS-7 cell cultures in assay buffer containing 50 ng/ml phorbol-12,13-dibutyrate (PdBu) (Sigma) and incubated at 37°C for 45 min. After gently washing three times with warm RPMI, adherent lymphoblasts were fixed with RPMI containing 1.5% (v/v) formalin. The numbers of lymphoblasts bound to COS-7 cells in multiple areas of formalin-fixed dishes were then counted using a Nikon Light microscope. To assess the inhibitory action of mAb, lymphoblasts or COS-7 cells were preincubated in medium containing mAb for 20 min before performing the binding assay in the continuous presence of the mAb.
Cloning of porcine ICAM-1 cDNA.
Initially we generated a porcine ICAM-1 cDNA fragment by RT-PCR, utilizing RNA obtained from PAEC stimulated with rhIL-1α and rhTNF-α for 4 hr and several primer pairs designed from the human and mouse ICAM-1 cDNA sequences. None of the primers based on the murine oligonucleotides yielded a product, and this was also the case for all but one of the human primer pairs. A 670-base pair (bp) PCR product was thus obtained using primers that spanned from position 598 bp to 1256 bp over domains 2–4 on the human sequence (E1: GG AGC CAA TTT CTC GTG CCG CAC and E2: CT CGT CCA GTC GGG GGC CAT ACA) (3). The sequence of this PCR product showed 71.8% and 63.4% identity to the comparable mouse and human ICAM-1 cDNA sequences. Using this sequence we generated two internal primers (I1: TGT TCC CAG CCT CAG AGG CTA C and I2: TGG TTC TTG TAT AGC ACA TGC C), which were used to create an authentic porcine ICAM-1 internal probe of 450 bp, and which were paired with T7 and T3 primers of pKS1 to amplify further porcine ICAM-1 sequences within a PAEC cDNA library (26). PCR products were gel purified and confirmed to be ICAM-1 by hybridization with the 450-bp internal probe and by sequencing. Together with relative sizing, this enabled the clones to be orientated with respect to the 670-bp probe, nucleotide sequences to be overlapped, and the sequence of the full-length cDNA to be constructed. Based upon this information, we were able to amplify full-length porcine ICAM-1 cDNA, again using the PAEC cDNA library as a template.
The porcine ICAM-1 cDNA we obtained was contained in a single 2440-bp clone and encoded an open reading frame of 1611 nucleotides (from positions 30–1641), corresponding to a protein of 537 amino acids.5 The predicted protein was predominantly hydrophilic, with each of five Ig-like extracellular domains consisting of approximately 100 amino acids and containing a total of 12 predicted N-linked glycosylation sites. The extracellular domains were followed by a 23-residue hydrophobic putative transmembrane domain and by a 29-residue putative cytoplasmic domain. Alignment of the porcine ICAM-1 cDNA with the human, mouse, rat, dog, and cow full-length homologues revealed regions of strong conservation in domains 1 and 3 involved in ligand binding. Domain 1 residues E34 and Q73 in the human sequence, which provide contact points for LFA-1 binding (31, 32), were completely conserved in all species, except for a Q73 to E substitution in the dog (Fig. 1A). Although the D229QR and E254DE motifs that have been implicated in Mac-1 binding to human domain 3 (6) were poorly conserved between species overall, the porcine sequence contained just one substitution in each motif, to DHR and EKE, respectively (Fig. 1B).
Porcine ICAM-1 can support CD18-dependent adhesion of human T lymphoblasts.
To determine whether porcine ICAM-1 was able to bind human LFA-1, an experiment was conducted examining the adhesion of phorbol ester-stimulated human T lymphoblasts to COS-7 cells that had been transfected with cDNA for either porcine ICAM-1 or porcine VCAM-1. As shown in Figure 2, lymphoblasts adhered to a comparable degree to ICAM-1 and VCAM-1 transfectants. The adhesion to porcine ICAM-1 transfectants was inhibited by mAb 6.5E (anti-CD18/β2 integrin) and not by mAb HP2/1 (anti-CD49d/α4 integrin), whereas the reverse was the case for adhesion to porcine VCAM-1 transfectants. Thus porcine ICAM-1 supports β2 integrin-mediated adhesion but not α4 integrin-mediated adhesion of human T lymphoblasts.
Generation of mAb to porcine ICAM-1.
Initial screening of a panel of mAb against human ICAM-1 failed to identify one that reacted with PAEC. Therefore to generate mAb against porcine ICAM-1, we immunized a mouse with PHA-activated porcine lymphoblasts. Hybridoma supernatants were screened using a cell-based ELISA for reactivity with TNF-α-stimulated PAEC. Positive clones were then rescreened by flow cytometry for binding COS-7 cells that had been transfected with porcine or human ICAM-1 cDNA. One of four antibodies selected by this approach, mAb 19C7 (IgM), was chosen for further study. As shown in Figure 3A, mAb 19C7 showed clear reactivity with COS-7 cells transfected with porcine ICAM-1 cDNA, but did not bind cells transfected with human ICAM-1 cDNA. Furthermore, mAb 19C7 significantly inhibited the adhesion of human lymphoblasts to porcine ICAM-1 COS-7 cell transfectants, but did not influence lymphoblast adhesion to human ICAM-1 COS-7 cell transfectants (Fig. 3B). Judging from the partial inhibition of lymphoblast adhesion that was observed, mAb 19C7 probably reacts with an epitope on porcine ICAM-1 that is not fully critical for adhesion. Attempts to use mAb 19C7 for Western blotting have so far been unsuccessful.
Expression of ICAM-1 on PAEC in response to cytokines.
The capacity of leukocytes to localize at inflammatory sites is to a large degree controlled by the actions of cytokines on EC adhesion molecule expression. Furthermore, our previous studies have shown that hrTNF-α and hrIL-1α are both capable of stimulating PAEC to increase expression of VCAM-1 (26). In an initial experiment, we used Northern analysis to determine whether porcine ICAM-1 expression by PAEC was induced by rhTNF-α. As shown in Figure 4, it was possible to detect constitutive expression in PAEC of mRNA for both ICAM-1 and VCAM-1. Levels of ICAM-1 steady-state mRNA were up-regulated between 2–4 hr after rhTNF-α stimulation, with the increase in VCAM-1 mRNA occurring slightly later between 4 and 8 hr. The level of steady-state mRNA for both molecules declined by 12 hr after stimulation with TNF-α.
Flow cytometry using a saturating concentration of mAb 19C7 was used to establish the basal expression of ICAM-1 and VCAM-1 and their up-regulation by hrTNF-α and hrIL-1α. Although there was some variation between EC cultures in the absolute levels of staining, it was notable that the staining of PAEC with anti-ICAM mAb was always less than that with anti-VCAM-1 mAb. Thus in nine cultures of unstimulated PAEC, the RFI of anti-ICAM-1 staining was 7.2±5.4 (mean ± SD), whereas that for anti-VCAM-1 mAb was 14.4±8.7. As shown in Figure 5, both ICAM-1 and VCAM-1 were gradually up-regulated by hrTNF-α, with maximal surface expression observed at 24 hr after stimulation. As with basal expression, the TNF-induced increase was more prominent for VCAM-1 (in five experiments the % increase over 24 hr of ICAM-1 was 119.4±40.5% and of VCAM-1 was 986±542.8%; both mean ± SD). In contrast to the effect of hrTNF-α, rhIL-1α led to a more rapid up-regulation of ICAM-1 expression, which was maximal at 6 hr and thereafter declined. Maximal VCAM-1 expression in response to rhIL-1α occurred slightly later at 12 hr. The apparently greater expression of VCAM-1 than ICAM-1 is a reversal of the relative expression of the two molecules by HUVEC, as seen in parallel cultures (unstimulated HUVEC: ICAM-1, RFI 10.5; VCAM-1, RFI 3.1; 24 hr TNF-α-stimulated HUVEC: ICAM-1, RFI 278.9; VCAM-1, RFI 68.0).
It is well established that IFN-γ alone stimulates expression of ICAM-1 by human EC, and that IFN-γ also synergistically enhances the effect of TNF-α on human EC ICAM-1 expression (34, 37, 38). We therefore tested the effects of recombinant IFN-γ preparations on PAEC. Human IFN-γ had no effect on ICAM-1 or VCAM-1 expression, consistent with the known species restriction of this cytokine (39, 40). Recombinant porcine IFN-γ weakly stimulated ICAM-1 expression when incubated alone with PAEC (RFI 11.0 unstimulated and 15.8 after 24 hr IFN-γ), but had an inhibitory effect on the increase in ICAM-1 due to TNF-α, both at 8 and 24 hr (Fig. 6A). rpIFN-γ alone also weakly stimulated VCAM-1 expression at 24 hr (RFI 35.0 unstimulated and 101.0 after 24 hr IFN-γ) but had no reproducible effect on the level of VCAM-1 induced by TNF-α (Fig. 6B).
In contrast to IFN-γ, IL-4 does not increase ICAM-1 expression by human EC but selectively augments the effect of TNF-α on VCAM-1 expression (41, 42). rhIL-4 had no reproducible effect on porcine ICAM-1 expression (Fig. 7A), but inhibited porcine VCAM-1 expression in response to TNF-α at 8 hr (Fig. 7B). This was in marked contrast to the augmenting action that the same preparation of IL-4 had on TNF-α-induced VCAM-1 expression by HUVEC in parallel cultures (not shown). Although the experiment shown in Figure 7 was conducted with human IL-4, identical results were obtained when PAEC were incubated with recombinant porcine IL-4 (not shown).
We have cloned cDNA and made mAb to porcine ICAM-1, thereby establishing the existence of this molecule in the pig. Although there has been indirect evidence for the presence of ligand(s) for β2 integrins on porcine EC (28–31), we had previously failed to identify existing anti-ICAM-1 mAb that reacted with a porcine homologue and had also failed to detect any hybridization of human ICAM-1 cDNA to porcine RNA upon Northern analysis. The cloning strategy that eventually led to our obtaining a full-length porcine ICAM-1 cDNA involved the generation of multiple PCR products, based upon the identification of a 670-bp fragment amplified by primers derived from human ICAM-1 cDNA. Porcine ICAM-1 is similar in structure to ICAM-1 in human and other species, with five Ig domains, a transmembrane region, and a cytoplasmic tail. The conservation of amino acid sequences is high throughout all five domains of porcine ICAM-1 compared with human ICAM-1, with the % identity being 51, 73, 47, 62, and 52 for domains 1–5, respectively. Interestingly, porcine ICAM-1 contains an RGD motif in the third Ig domain, at amino acid positions 258–260, located immediately proximal to a D260HR motif that may be equivalent to the D229QR Mac-1 binding motif in human ICAM-1 (6). Whether this RGD sequence influences the function of porcine ICAM-1 remains to be determined.
By generating an mAb to porcine ICAM-1 we were able to examine the expression pattern of this molecule on porcine EC. The low level of binding of anti-ICAM-1 mAb 19C7 to PAEC is unlikely to have been due to a low affinity, because the antibody reacted well with porcine ICAM-1 COS-7 cell transfectants. Moreover, in unpublished work we have found that the anti-ICAM-1 mAb 19C7 reacts well with unstimulated and TNF-α-stimulated porcine vascular smooth muscle cells (TY Huehns, CJ Stocker, and DO Haskard, unpublished observations). Thus, although possible differences in binding characteristics between the two antibodies make it difficult to make precise statements about the relative density of expression of ICAM-1 and VCAM-1, our data suggest that expression of ICAM-1 on unstimulated and on cytokine-activated PAEC is at a relatively low level compared with that of VCAM-1. This is the reverse of the situation with HUVEC and may account, together with differences in molecular conservation between species, for the experience that anti-VCAM-1 mAb tend to predominate over anti-ICAM-1 when hybridomas are generated from mice immunized with cytokine-activated PAEC and vice versa when mice are immunized with cytokine-activated HUVEC. Whether the difference between PAEC and HUVEC is related to the species or the tissue will await a direct comparison between porcine and human aortic EC.
Besides showing differences between PAEC and HUVEC in levels of basal expression, ICAM-1 and VCAM-1 also show subtle differences between the two cell types in cytokine responsiveness. This was most noticeable when IFN-γ or IL-4 were combined with TNF-α. Although with human EC the combinations of IFN-γ + TNF-α and IL-4 + TNF-α tend to synergistically enhance ICAM-1 and VCAM-1 expression, respectively (37, 38, 41, 42), the effects of these combinations on PAEC were to reduce the effects of TNF-α on the respective adhesion molecules. Further functional experiments will be needed to establish the exact significance of these findings in terms of leukocyte adhesion, but the data suggest that IFN-γ and IL-4 may have inhibitory effects on the expression of integrin ligands on PAEC, as has been reported with human EC for E-selectin (43). In the case of xenograft rejection, it is therefore possible that IFN-γ and IL-4 may serve to reduce leukocyte adhesion to endothelium. Our data suggest that the expression by porcine endothelium of ICAM-1 and VCAM-1 would be influenced by IL-4 or IFN-γ released by porcine passenger cells within the graft, or by IL-4 released by infiltrating human lymphocytes. On the other hand, human IFN-γ is not likely to influence porcine EC adhesion molecule expression directly.
Although LFA-1 of one species does not always bind ICAM-1 from another (44), we found that human lymphoblasts bound COS-7 cells transfected with porcine ICAM-1 by an LFA-1-dependent mechanism. Since it is already known that human VLA-4 binds porcine VCAM-1 (45), our findings indicate that human host responses to porcine xenografts are unlikely to be impaired by species incompatibilities involving these reciprocal adhesion mechanisms. Conversely, mAb or other agents that specifically target porcine ICAM-1 or VCAM-1 could be an effective strategy for inhibiting leukocyte adhesion to xenograft endothelium.
1. Rothlein R, Dustin ML, Marlin SD, Springer TA. A human intercellular adhesion molecule (ICAM-1) distinct from LFA-1. J Immunol 1986; 137: 1270.
2. Marlin SD, Springer TA. Purified intercellular adhesion molecule 1 (ICAM-1) is a ligand for lymphocyte function associated antigen 1 (LFA-1). Cell 1987; 51: 813.
3. Staunton DE, Marlin SD, Stratova C, Dustin ML, Springer TA. Primary structure of ICAM-1 demonstrates interaction between members of the immunoglobulin and integrin supergene families. Cell 1988; 52: 925.
4. Makgoba MW, Sanders ME, Ginther Luce GE, et al. ICAM-1 a ligand for LFA-1-dependent adhesion of B, T, and myeloid cells. Nature 1988; 331: 86.
5. Diamond MS, Staunton DE, De Fougerolles AR, et al. ICAm-1 (CD54): A counter-receptor for Mac-1 (CD11b/CD18). J Cell Biol 1990; 111: 3129.
6. Diamond MS, Staunton DE, Marlin SD, Springer TA. Binding of the integrin Mac-1 (CD11b/CD18) to the third immunoglobulin-like domain of ICAm-1 (CD54) and its regulation by glycosylation. Cell 1991; 65: 961.
7. Languino LR, Plescia J, Duperray A, et al. Fibrinogen mediates leukocyte adhesion to vascular endothelium through an ICAm-1-dependent pathway. Cell 1993; 73: 1423.
8. Manning AM, Lu HF, Kukielka GL, et al. Cloning and comparative sequence analysis of the gene encoding canine intercellular adhesion molecule-1. Gene 1995; 156: 291.
9. Kita Y, Takashi T, Iigo Y, Tamatani T, Miyasaka M, Horiuchi T. Sequence and expression of rat ICAM-1. Biochim Biophys Acta Gene Struct Expression 1992; 1131: 108.
10. Siu G, Hedrick SM, Brian AA. Isolation of the murine intercellular adhesion molecule 1 (ICAM-1) gene. ICAM-1 enhances antigen-specific T cell activation. J Immunol 1999; 143: 3813.
11. Ballantyne CM, O’Brien WE, Beaudet AL. Nucleotide sequence of the cDNA for murine intercellular adhesion molecule-1 (ICAM-1). Nucleic Acids Res 1989; 17: 5853.
12. Lee E, Kang SG, Kehrli MEJr . Cloning, sequencing and analysis of cDNA encoding bovine intercellular adhesion molecule-1 (ICAM-1). Vet Immunol Immunopathol 1997; 59: 121.
13. Springer TA. Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration. Annu Rev Physiol 1995; 57: 827.
14. Cosimi AB, Conti D, Delmonico FL, et al. In vivo effects of monoclonal antibody to ICAM-1 (CD54) in nonhuman primates with renal allografts. J Immunol 1990; 144: 4604.
15. Isobe M, Yagita H, Okumura K, Ihara A. Specific acceptance of cardiac allograft after treatment with antibodies to ICAM-1 and LFA-1. Science 1992; 255: 1125.
16. Taylor PM, Rose ML, Yacoub MH, Pigott R. Induction of vascular adhesion molecules during rejection of human cardiac allografts. Transplantation 1992; 54: 451.
17. Orosz CG, Ohye RG, Pelletier RP, et al. Treatment with anti-vascular cell adhesion molecule 1 monoclonal antibody induces long-term murine cardiac allograft acceptance. Transplantation 1993; 56: 453.
18. Isobe M, Suzuki J, Yagita H, et al. Immunosuppression to cardiac allografts and soluble antigens by anti-vascular cellular adhesion molecule-1 and anti-very late antigen-4 monoclonal antibodies. J Immunol 1994; 153: 5810.
19. Wang CY, Naka Y, Liao H, et al. Cardiac graft intercellular adhesion molecule-1 (ICAM-1) and interleukin-1 expression mediate primary isograft failure and induction of ICAM-1 in organs remote from the site of transplantation. Circ Res 1998; 82: 762.
20. Osborn L, Hession C, Tizard R, et al. Direct expression cloning of vascular cell adhesion molecule 1, a cytokine-induced endothelial protein that binds to lymphocytes. Cell 1989; 59: 1203.
21. Jones DA, McIntyre LV, Smith CW, Picker LJ. A two-step adhesion cascade for T cell/endothelial cell interactions under flow conditions. J Clin Invest 1994; 94: 2443.
22. Alon R, Kassner PD, Woldemar Carr M, Finger EB, Hemler ME, Springer TA. The integrin VLA-4 supports tethering and rolling in flow on VCAM-1. J Cell Biol 1995; 128: 1243.
23. Pober JS, Cotran RS. Cytokines and endothelial cell biology. Physiol Rev 1990; 70: 427.
24. Bach FH. Xenotransplantation. Annu Rev Med 1999; 49: 301.
25. Auchincloss H, Sachs DH. Xenogeneic transplantation. Annu Rev Immunol 1999; 16: 433.
26. Tsang YTM, Haskard DO, Robinson MK. Cloning and expression kinetics of porcine vascular cell adhesion molecule. Biochem Biophys Res Commun 1994; 201: 805.
27. Harrison AA, Stocker CJ, Chapman PT, et al. Expression of VCAM-1 by vascular endothelial cells in immune- and non-immune inflammatory reactions in the skin. J Immunol 1997; 159: 4546.
28. Pleass HC, Forsythe JL, Proud G, Taylor RM, Kirby JA. Xenotransplantation: an examination of the adhesive interactions between human lymphocytes and porcine renal epithelial cells. Transpl Immunol 1994; 2: 225.
29. Robinson LA, Tu L, Steeber DA, Preis O, Platt JL, Tedder TF. The role of adhesion molecules in human leukocyte attachment to porcine vascular endothelium: implications for xenotransplantation. J Immunol 1999; 161: 6931.
30. Rollins SA, Kennedy SP, Chodera AJ, Elliott EA, Zavoico GB, Matis LA. Evidence that activation of human T cells by porcine endothelium involves direct recognition of porcine SLA and costimulation by porcine ligands for LFA-1 and CD2. Transplantation 1994; 57: 1709.
31. Herrlinger KR, Eckstein V, Muller-Ruchholtz W, Ulrichs K. Human T-cell activation is mediate predominantly by direct recognition of porcine SLA and involves accessory molecule interaction of ICAM1/LFA 1 and CD2/LFA3. Transplant Proc 1996; 28: 650.
32. Horner A, Davies ME, Franz B. Chondrocyte-peripheral blood mononuclear cell interactions: the role of ICAM-1. Immunology 1995; 86: 584.
33. Thornhill M, Kyan-Aung U, Lee TL, Haskard DO. T cells and neutrophils exhibit differential adhesion to cytokine stimulated endothelial cells. Immunology 1990; 69: 287.
34. Wellicome SM, Thornhill MH, Pitzalis C, et al. A monoclonal antibody that detects a novel antigen on endothelial cells that is induced by tumor necrosis factor, IL-1 or lipopolysaccharide. J Immunol 1990; 144: 2558.
35. McLean IW, Nakane PK. Periodate-lysine-paraformaldehyde fixative: a new fixative for immunoelectron microscopy. J Histochem Cytochem 1974; 22: 1077.
36. Sambrook J, Fristch EF, Maniatis T. Molecular cloning: a laboratory manual. New York: Cold Spring Harbor Laboratory Press, 1989.
37. Pober JS, Gimbrone MA, Lapierre LA, et al. Overlapping patterns of activation of human endothelial cells by interleukin 1, tumour necrosis factor, and immune interferon. J Immunol 1986; 137: 1893.
38. Dustin ML, Springer TA. Lymphocyte function-associated antigen-1 (LFA-1) interaction with inter-cellular adhesion molecule-1 (ICAM-1) is one of at least three mechanisms for lymphocyte adhesion to cultured endothelial cells. J Cell Biol 1988; 107: 321.
39. Murray AG, Khodadoust MM, Pober JS, Bothwell ALM. Porcine aortic endothelial cells activate human T cells: direct presentation of MHC antigens and costimulation by ligands for human CD2 and CD28. Immunity 1994; 1: 57.
40. Sultan P, Murray AG, McNiff JM, et al. Pig but not human interferon-γ in vivo initiates human cell-mediated rejection of pig tissue in vivo. Proc Natl Acad Sci USA 1999; 94: 8767.
41. Thornhill MH, Haskard DO. IL-4 regulates endothelial activation by IL-1, tumor necrosis factor or IFNγ. J Immunol 1990; 145: 865.
42. Iademarco MF, Barks JL, Dean DC. Regulation of vascular cell adhesion molecule-1 expression by IL-4 and TNF-α in cultured endothelial cells. J Clin Invest 1995; 95: 264.
43. Melrose J, Tsurushita N, Liu G, Berg EL. IFN-gamma inhibits activation-induced expression of E- and P-selectin on endothelial cells. J Immunol 1998; 161: 2457.
44. Johnston SC, Dustin ML, Hibbs ML, Springer TA. On the species specificity of the interaction of LFA-1 with intercellular adhesion molecules. J Immunol 1990; 145: 1181.
45. Dorling A, Stocker C, Tsao T, Haskard D, Lechler RI. In vitro accommodation of immortalized porcine endothelial cells: resistance to complement mediated lysis and down-regulation of VCAM expression induced by low concentrations of polyclonal human IgG anti-pig antibodies. Transplantation 1996; 62: 1127.