| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2003;125:306-314
© 2003 The American Association for Thoracic Surgery
Surgery for Acquired Cardiovascular Disease (ACD) |
1-3Galß1-4GlcNAc-R (Gal) expression on porcine valve endothelial cells
From the Department of Surgery, Brigham and Women's Hospital, Boston, Mass,a and the Department of Cardiothoracic Surgery, Mount Sinai Medical Center, New York, NY.b
Received for publication Dec 26, 2001. Revisions requested March 11, 2002; revisions received April 19, 2002. Accepted for publication May 1, 2002. Address for reprints: David H. Adams, MD, Department of Cardiovascular and Thoracic Surgery, Mount Sinai Medical Center, 1190 Fifth Ave, New York, NY 10029 (E-mail: david.adams{at}mountsinai.org).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
1-3Galß1-4GlcNAc-R (known as Gal), suggesting that fresh porcine valve may be immunoprivileged. In this study, we further investigated the mechanisms of Gal expression on porcine valve endothelial cells. Gal at both the protein and messenger RNA levels. Gal much less intensely than did aortic endothelial cells. Porcine aortic endothelial cells expressed an isolectin B4 (isolectin B4 lectin Bandeiraea simplicifolia) immunodetectable band at 135 kd that was not visible on porcine valve endothelial cells or on human vein endothelial cells. Reverse transcriptase-polymerase chain reaction documented three transcripts of the Gal gene that were identically expressed on porcine valve and aortic endothelial cells. Furthermore, flow cytometry showed an almost identical surface profile between porcine aortic and valve endothelial cells, in contrast with human vein endothelial cells. Gal, although whole-cell extracts demonstrated an immunodetectable band on Western blotting that was present on aortic endothelial cells but not on valve endothelial cells. These findings clarify the mechanism of expression of 1,3galactosyltransferase gene expression in valve endothelial cells, suggesting that delayed rejection of fresh porcine cardiac valves may occur. |
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
1 It should provide excellent hemodynamic properties, provide durability, and have a minimal need for anticoagulation.
The development of a long-term viable bioprosthesis has been hampered by structural deterioration when placed into younger patients needing valve replacement. 2 It has been theorized that a viable bioprosthesis would contain regenerating fibroblasts and valve endothelial cells to provide a turnover of matrix and antithrombotic factors. Fibroblasts could produce collagen to stiffen the underlying valve matrix. Valvular endothelial cells could provide the proper balance between procoagulant and anticoagulant forces to minimize the need for long-term anticoagulation. The hemodynamic benefits of bioprostheses might be realized without the long-term structural deterioration that some have linked to glutaraldehyde fixation. Concern has been expressed that a viable endothelial surface displays immunologically xenoreactive epitopes, such as major histocompatibility complex class II, intracellular adhesion molecule 1, CD54, and E-selectin, on the surface of valve endothelium. 3,4 For example, some centers have demonstrated that ABO mismatch is an independent predictor of graft deterioration, and typing thus may improve the long-term outcomes of homograft valves, 5 although this is controversial. 6
The interaction between primate xenoreactive antibodies and the carbohydrate Gal1-3Galß1-4GlcNAc-R (Gal) is the major acute obstacle faced when xenotransplanting porcine tissue. 7 The Gal epitope is synthesized by the glycosylation enzyme 1,3galactosyltransferase. The Gal epitope is homologous to the ABO blood group antigens; in effect, pigs have an additional blood group antigen.
We have demonstrated that native porcine cardiac valves do not have any immunohistochemically detectable Gal on the surface, as shown by immunostaining with isolectin B4 (IB4) lectin. 8 Furthermore, when hearts from transgenic swine expressing human complement regulatory proteins (CD59 and decay-accelerating factor) are implanted into baboons, the valves, unlike the myocardium, are spared from immunoglobulin M and membrane attack complex deposition, thought to be initiated by the expression and binding of xenoreactive antigens to endothelially expressed Gal. 9 We later found that freshly harvested human cardiac valves have no immunohistochemically detectable ABO blood group antigens, which are the human homologs of Gal, on their surface. 10 We then attempted to use cell culture to define the surface epitopes on human vein endothelial cells and human saphenous vein endothelial cells. In culture, they appeared to lack ABH reactivity, 6,11 although they have been shown to stimulate human T-cell proliferation and interleukin 2 release. 3
Because saphenous vein endothelial cells may not be a good proxy for valve endothelial cells, we established cultures of porcine aortic endothelial cells and porcine valve endothelial cells and applied Western blotting, immunohistochemical methods, RNA analysis, and flow cytometry to aortic and valve endothelial cells to investigate further the mechanism of Gal expression.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-actin (A-2547); diaminobenzidine; penicillin G, streptomycin, and glutamine solutions, trypsin-EDTA solution for endothelial cells; and Tri-Reagent were purchased from Sigma (St Louis, Mo). Oligonucleotide primers for human ß-actin were purchased from Stratagene (LaJolla, Calif). Oligonucleotides created for reverse transcriptase-polymerase chain reaction (PCR) for Gal were purchased from Sigma Genosys (Woodlands, Tex). Precast agarose gels (4%), prestained molecular weight markers, and precut 0.20-µm polyvinylidene difluoride membranes were purchased from Invitrogen Corporation (Carlsbad, Calif). Casein (I-block) was purchased from Tropix (Bedford, Mass). Endothelial cell growth factor was purchased from (Roche Diagnostics Corporation, Laboratory Systems (Indianapolis, Ind). Human vein endothelial cells (catalog CRL-1730) and Kaighn F-12 medium were purchased from American Type Culture Collection (Manassas, Va). Primary cultures of porcine aortic endothelial cells were ordered from Cell Systems (Kirkland, Wash). The Vectastain Elite avidin-biotin complex kit was purchased from Vector Laboratories, Inc (Burlingame, Calif). Superscript II reverse transcriptase, medium 199 with Earle salts, Dulbecco phosphate-buffered saline solution lacking calcium or magnesium, and gentamicin were purchased from Life Technologies, Inc (Rockville, Md). Fetal calf serum (heat inactivated) was purchased from HyClone Laboratories Inc (Logan, Utah). Adult 50-kg female pigs were purchased from Parson Farms (Hadley, Mass). Sodium pentobarbital was purchased from Delmarva Laboratories (Midlothian, Va). Six-well tissue culture plates were purchased from Denville Scientific (Metuchen, NJ). Crystal/Mount was purchased from Biomeda Corp (Foster City, Calif).
Immunohistochemical studies
Cells were grown to near confluence on plastic tissue culture-treated slides. Cells were fixed for 10 minutes in ice-cold 10% methanol and phosphate-buffered saline solution. For lectins, a 1:100 solution of biotinylated IB4 lectin or RCA lectin was applied for 1 hour at room temperature. Slides were then developed by incubating for 30 minutes in avidin-biotin complex, which was then developed with diaminobenzidine. Washes for lectins were done with phosphate-buffered saline washes without polysorbate (Tween) 20. The high stringency of Tween 20 resulted in no staining (data not shown). Counterstaining was with Gill hematoxylin for 1 minute. Slides were then mounted with an aqueous mount. For smooth muscle cell -actin immunostaining, primary antibody was used at a concentration of 1:10,000 in phosphate-buffered saline solution and 1% bovine serum albumin for 1 hour. Washes were all performed in phosphate-buffered saline solution with Tween (0.5%). Secondary antibody incubation (1:2000 in phosphate-buffered saline solution with Tween) was for 1 hour at room temperature. Avidin-biotin complex reagent was then applied for 1 hour, and diaminobenzidine was used as a substrate for the horseradish peroxidase reaction.
Sections of tissue were used as a control for antibodies. Sections were cut 4-µm thick on a Leica microtome (Leica Microsystems Inc, Deerfield, Ill) from formalin-fixed, paraffin-embedded tissues. Standard deparaffinization and dehydration were performed. Tissue was treated with protease K for 10 minutes. Incubation in avidin-biotin complex reagent alone served as a nonspecific negative control preparation.
Cell culture: Porcine valvular endothelial cells
Primary cultures of porcine valvular endothelial cells were prepared with the following modifications of previously published methods. Hearts were removed under aseptic conditions from domestic adult swine (50 kg) that were killed by injection of 87 mg/kg sodium pentobarbital. All animals were treated humanely in compliance with the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources, National Research Council, and published by the National Academy Press, revised 1996. The valve leaflets were removed from the aortic and mitral valves and placed into phosphate-buffered saline solution supplemented with antibiotics (100 U/mL penicillin G and 20 µg/mL gentamicin). A margin of leaflet was left attached to the anulus. Valve leaflets were washed free of red blood cells by three washes with phosphate-buffered saline and antibiotic solution. The valve leaflets were then placed in a 0.6-mg/mL collagenase solution (from C hystoliticum) in phosphate-buffered saline solution supplemented with 100 U/mL penicillin G and 20 µg/mL gentamicin at 37°C for 15 minutes. The valve leaflet was then gently shaken for 1 minute in medium 199. Cells that became dislodged were then placed by aliquots into a 96-well plate into which complete medium was added. Medium was changed every 48 hours. Cells were checked for expansion every day. Cells with fibroblastic or smooth muscle cell morphologic and growth characteristics, as evinced by spindle morphologic characteristics or lack of contact inhibition, were discarded. Cells of typical endothelial morphologic type were then expanded and confirmed to be of endothelial cell origin by immunohistochemical methods. These cells were used for subsequent experiments. Complete medium consisted of medium 199 with Earle salts supplemented with 20% heat-inactivated fetal calf serum, penicillin at 100 U/mL, streptomycin at 0.1 mg/mL, l-glutamine at 2 mmol/L, endothelial cell growth factor at 100 µg/mL, and heparin at 50 µg/mL. Cells were subcultured at near confluence with a standard trypsin (500 BAEE) and EDTA (180 µg/mL) detachment procedure.
Porcine aortic endothelial cells were obtained during first passage (<6 cumulative population doublings) and routinely used for experiments between passages 4 through 6. They were isolated from the aortas of large white pigs, as described by Ryan and Maxwell. 12 They were grown in complete medium 199, which was changed every 48 to 72 hours. Human vein endothelial cells were grown in Kaighn's F12 medium supplemented with 10% heat-inactivated fetal calf serum, penicillin (100 U/mL), streptomycin (0.1 mg/mL), and l-glutamine (2 mmol/L), endothelial cell growth factor (100 µg/mL), and heparin (50 µg/mL). Cells were subcultured near confluence with trypsin.
Reverse transcriptase-polymerase chain reaction
RNA was extracted with previously described modifications 13 of the method of Chomczynski and Sacchi. 14 In brief, endothelial cells were lysed with 4 mL Tri-Reagent per 10-cm petri dish of confluent cells. Chloroform was added at one fifth final volume and the solution was vigorously vortex mixed. Nucleoprotein complexes were then allowed to dissolve by letting the homogenate sit for 15 minutes on ice. The solution was centrifuged for 30 minutes at 4°C and 12000g. The aqueous layer was then removed and mixed with an equal volume of isopropanol. The solution was precipitated overnight at -20°C. Subsequently, the isopropanol and RNA mixture was centrifuged at 12000g for 30 minutes at 4°C. The resulting RNA pellet was washed with a total of 6 mL of 70% ethanol and centrifuged again at 7500g at 4°C for 15 minutes. The RNA pellet extracted from cell culture was dissolved in 40 µL of water. RNA was quantified by absorbence at 260 nm, with a 260 nm/280 nm ratio of 1.7 to 1.9. All experiments were performed in triplicate with identical results, and a single representative autoradiograph is shown.
Reverse transcription was performed according to the Superscript II directions. In brief, 2 µg of total RNA was mixed with oligodeoxythymidine primer (0.5 µg/µL) and annealed for 10 minutes at 70°C and then placed on ice. PCR buffer (20-mmol/L tris[hydroxymethyl]aminomethane [Tris, pH 8.4] and 50-mmol/L potassium chloride), magnesium chloride (2.5 mmol/L), deoxyribonucleoside triphosphate (1 mmol/L), and dithiothreitol (10 µmol/L) were added. Samples were heated to 42°C for 5 minutes, at which point reverse transcriptase was added to the mixture. The reaction was allowed to proceed for 50 minutes and then terminated by 5 minutes of heating to 94°C. Five percent (1 µL) of this reaction was routinely used for PCR amplification.
The human ß-actin primer consisted of the sense sequence 5'-TGA CGG GGT CAC CCA CAC TGT GCC CAT CTA-3' and the antisense sequence 5'-CTA GAA GCA TTT GCG GTG GAC GAT GGA GGG-3', resulting in a 661-base pair (bp) fragment. 15 Reverse transcription of the RNA extracts was performed from 2 µg of total RNA with oligodeoxythymidine as initiation primer. PCR amplification of a 5' fragment of the pig Gal coding region was done with the following primers: 5'-AGG AAG AGT GGT TCT GTC-3', corresponding to nucleotides 12 through 30 of pig 1,3GT isoform 1, and 5'-GTT ATG GTC ACG ACC TCT-3', corresponding to nucleotides 324 through 306 of pig 1,3GT isoform 1, as previously described elsewhere. 16,17 PCR reaction mix consisted of 100 pmol primers, 10 mmol/L-deoxyribonucleoside triphosphates, 3-mmol/L magnesium chloride, 10-mmol/L Tris hydrochloride buffer, 50-mmol/L potassium chloride, 10% glycerol, and 2 U Taq polymerase. Conditions for PCR reaction were as follows: 94°C denaturing temperature for 30 seconds, 54°C hybridization for 30 seconds, and 72°C elongation temperature. PCR was performed for 35 cycles. PCR reactions were resolved on precast 4% agarose gels and stained with ethidium bromide. Images were recorded with Polaroid 667 film (Polaroid Corporation, Cambridge, Mass).
Western blot
Porcine aortic endothelial cells, porcine valve endothelial cells, and human vein endothelial cell monolayers were washed with phosphate-buffered saline solution containing 5-mmol/L EDTA, scraped from the flasks in phosphate-buffered saline solution and EDTA, and centrifuged at 900g for 5 minutes at 4°C in a swinging bucket centrifuge. Cell pellets were then lysed in 1 volume of lysis buffer (50-mmol/L Tris [pH 7.2] containing 2% Triton X-100 [The Dow Chemical Company, Midland, Mich], 5-mmol/L EDTA, and protease inhibitors [pepstatin A, leupeptin, bestatin, aprotinin, and E-64]) for 1 hour on ice. After sedimentation of nuclei and debris at 10,000g for 30 minutes, supernatants were collected and stored at -80°C until use. Protein concentration was determined by the bicinchoninic acid method, and 20 µg of total protein per lane was analyzed by immunoblotting. Aliquots of endothelial cell extracts were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis on an 8% gel under reducing conditions according to the method of Laemmli. 18 A second gel was run as a control to ascertain equal loading and transfer, which was confirmed by staining with Coomassie blue. Proteins were then transferred to a 0.45-µmol/L polyvinylidene difluoride membrane at 125 V for 2 hours and stained with biotinylated IB4 lectin, according to minor modifications of a previously described method. 19 In brief, biotinylated IB4 lectin was used at a 1:100 concentration (diluted in phosphate-buffered saline solution) for 1 hour at room temperature after blocking overnight with 0.2% casein and Tris-buffered saline Tween (0.05%) solution at 4° C. The membrane was then washed three times for 5 minutes each in phosphate-buffered saline solution. The Vectastain Elite kit was then used according to directions to develop the immunoblot. It was critical to eliminate Tween from the wash and incubation buffers, because the IB4 lectin displays no signal in the presence of stringent wash conditions (data not shown). Control blots were run in the absence of lectin and stained with 3,3'-diaminobenzidine with virtually no background signal. As an additional control, blots included a lane of human vein endothelial cell total protein extract, which is known not to express Gal. All experiments were performed in triplicate, with a single representative blot being shown.
Flow cytometry
Cells were allowed to grow to near confluence and then harvested with trypsin, which was neutralized in phosphate-buffered saline solution with 1% bovine serum albumin, washed twice with phosphate buffered-saline solution, and resuspended in phosphate-buffered saline solution with 1% bovine serum albumin. Flow cytometric analysis was performed according to minor modifications of previously published protocols 20 by incubating 100 µL (2 x 105 cells) of porcine aortic or valve endothelial cells or human vein endothelial cells with 100 µL of fluorescein isothiocyanate-conjugated IB4 lectin (20 µg/mL) in phosphate-buffered saline solution with 0.5-mmol/L calcium chloride for 30 minutes on ice. After double washing with phosphate-buffered saline with 0.5-mmol/L calcium chloride, the cells were resuspended in phosphate-buffered saline solution and analyzed on an EPICS-XL flow cytometer (Coulter Corporation, Hialeah, Fla) at 488 nm. Human vein endothelial cells, which are known not to express Gal, served as a negative control for the IB4 lectin. All experiments were performed twice with one representative scan shown.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
21,22 initial experiments were established to determine the proper growth parameters. Cells grew rapidly when placed on a 2% bovine gelatin substrate but not on standard cell culture dishes (data not shown). Cells had typical cobblestone morphologic character and exhibited contact inhibition in the presence of 20% fetal calf serum (data not shown).
Immunohistochemical studies
Cells were initially characterized by staining with the IB4 lectin to determine Gal reactivity. Whereas porcine aortic endothelial cells (Figure 1, A) stained intensely with IB4 lectin, porcine valve endothelial cells (Figure 1, B) showed minimal immunoreactivity for the IB4 lectin. Porcine myocardium (Figure 1, C) served as both positive and negative tissue controls for the lectin. Porcine myocardium stained intensely in the capillaries and arterioles for Gal, whereas cardiomyocytes were completely unstained and larger epicardial vessels showed minimal reactivity. Cells were routinely photographed at 450x magnification and tissue sections were routinely photographed at 250x magnification.
|
23 Both aortic (Figure 1
, D) and valve (Figure 1, E) endothelial cells stained positively for RCA, confirming their endothelial origin. As a control, myocardium was stained, as shown in Figure 1, F, with the RCA lectin. Endothelial cells, both microvascular and large epicardial vessels, appeared to stain intensely with this lectin, which is consistent with previously published reports. 24
As further confirmation, immunohistochemical testing was performed for smooth muscle cell -actin. Both aortic (Figure 1, G) and valve (Figure 1, H) endothelial cells had minimal reactivity toward smooth muscle cell -actin. As a staining control, porcine myocardium was stained with the same antibody (Figure 1, I). There was appropriate staining in the smooth muscle layers of both arteries and veins.
In summary, valvular endothelial cells in culture exhibited the same IB4 staining patterns (RCA positive, Gal negative) in vitro as in vivo with respect to endothelial cell phenotype.
Western blot
Whole-cell homogenates were obtained for analysis of IB4 by immunoblotting. Twenty micrograms of total protein from three different experiments was loaded for both porcine valve and aortic endothelial cells. Human vein endothelial cells were included as a control, and a molecular weight marker was used. There was an intensely positive Gal band in aortic endothelial cells at 135 kd (Figure 2); this band was lacking in porcine valve endothelial cells and human vein endothelial cells. A less intense band was evident at 125 kd in valve endothelial cells.
|
Gal. Three amplified products were visualized for the Gal gene in both porcine aortic and valve endothelial cells (Figure 3). These corresponded to previously published sizes and distribution patterns. 17 Although reverse transcriptase-PCR is not an absolutely quantifiable method but rather a relative one, it was obvious that both samples contained similar amounts of amplified product, in the same sizes, with similar amounts of ß-actin, a constitutive gene that was coamplified. Sequencing revealed that base pairs were 286, 313, and 349 nucleotides in length. Films were scanned and intensity was measured. In contrast to our findings with immunohistochemical and Western blot analyses, there were no significant differences in band intensity between transcripts from aortic and valve endothelial cell messenger RNA.
|
Gal and served as a negative control; Figure 4).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Gal epitope on porcine valve endothelial cells. Gal is the major xenoreactive epitope in the pig and as such is the antigen responsible for hyperacute rejection of porcine xenografts. We have previously shown that there is no immunohistochemically detectable Gal on valve endothelial cells, although there is intense Gal staining in capillaries. 8,25 As such, valve endothelium may represent immunologically privileged tissue, suggesting a potential role for a fresh cardiac valve prosthesis.
Results of this investigations shed new light on the apparent lack of Gal epitopes on cardiac valve endothelium. Whereas immunohistochemical and Western blot analyses documented decreased Gal reactivity in valve endothelial cells, flow cytometry and reverse transcriptase-PCR demonstrated nearly equivalent surface epitope and transcript profiles for Gal, implying that immunostaining with the IB4 lectin may be ineffective in identifying relevant Gal epitopes.
First, we established a method for the primary cultures of valve endothelial cells. We documented that cultured valve endothelial cells appeared to have similar phenotypic expression to that on valve leaflets. For example, cultured valve endothelial cells exhibited typical cobblestone morphologic type and were RCA lectin positive, confirming their endothelial lineage. 23 Furthermore, in contrast to porcine aortic endothelial cells, valve endothelial cells expressed almost no Gal according to IB4 immunostaining. This finding is similar to our previous reports, 8,9 in which we documented that there was no IB4 detectable staining on the endothelium of porcine cardiac valves before or after xenotransplantation into primates. We were also able to confirm by Western blot analyses that valve endothelial cells expressed less IB4-detectable Gal than did aortic endothelial cells. This is in good agreement with our previous histologic results.
Reverse transcriptase-PCR demonstrated Gal gene expression on both aortic and valvular endothelial cells, also demonstrating equivalent amounts of transcript in the two cell types. The 1,3galactosyltransferase gene is alternatively spliced from a single genomic locus that synthesizes four proteins with variations in the length of the Golgi-spanning stem region. These four isoproteins have all been shown to have 1,3galactosyltransferase activity, with four possible transcripts of 250, 286, 313, and 349 bp. 26 Here we demonstrated three transcripts, which is consistent with previously published reports. 17 This is the first report to our knowledge of the differential display of these transcripts in porcine valve endothelial cells. When compared with a constitutive gene, such as the ß-actin gene, there are equal amounts of all three transcripts.
Finally, flow cytometry with the IB4 lectin revealed identical surface profiles in aortic and valve endothelial cells, which were distinct from human vein endothelial cells. This is consistent with the reverse transcriptase-PCR activity of these cells. Apparently the IB4 lectin did not have sufficient sensitivity in the harsh environment of histochemical and Western washes to detect low levels of Gal epitope expressed on the endothelial cell surface. In the more gentle conditions of flow cytometry required for lectin-Gal interactions, it is possible that there is a tighter IB4-Gal interaction.
Most studies investigating the expression of Gal have used the IB4 lectin from B (Griffonia) simplicifolia for immunohistochemical 23,27,28 or fluorescence-activated cell sorter analyses. 20 This lectin is from a family of tetravalent lectins that bind Gal, with the IB4 lectin completely specific for Gal. Although its specificity is high, the sensitivity of the IB4 lectin may be insufficient for the detection of small numbers of the Gal epitope. There is a low binding affinity of each monomeric interaction of the IB4 lectin with Gal. Because of this, it has been postulated that if the Gal epitopes are not in close enough proximity to enable a IB4-Gal interaction with more than one molecule of the lectin (which has four combining sites), then the lectin may be removed during overly stringent washes. 29,30 This may explain why, despite messenger RNA and fluorescence-activated cell sorter analyses demonstrating transcript and IB4-detectable Gal epitopes, respectively, there is minimal staining by immunohistochemical means. Also, posttranscriptional modifications of glycoproteins or differences in Gal presentation have been noted to affect the relative abundance of immunodetectable Gal. 31
Other tissue beds that have been thought to be immunologically privileged have been shown to have low-level expression of Gal epitopes. Indeed, it has been demonstrated that despite a low expression of Gal epitopes in bovine and porcine cartilage, there is a marked (20- to 100-fold) increase in anti-Gal activity when cartilage is placed into suprapatellar pouches in cynomolgus monkeys. 32,33 This emphasizes the need to remove all Gal epitopes, because even a very low expression may be sufficient to propagate a chronic rejection cascade.
This study confirms that porcine valve endothelium expresses Gal epitopes in vitro, although these epitopes are not detectable by immunohistochemical means. Furthermore, equivalent amounts of 1,3galactosyltransferase messenger RNA isoforms are present in valve and aortic endothelial cells, suggesting that any differences may be due to posttranscriptional modifications or differences in antigen presentation. Unless all Gal epitopes are removed from the endothelium, such as by targeted gene disruption, our results suggest that fresh porcine valves are ultimately likely to trigger delayed xenograft rejection.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. S. Farivar, L. H. Cohn, E. G. Soltesz, T. Mihaljevic, J. D. Rawn, and J. G. Byrne Transcriptional profiling and growth kinetics of endothelium reveals differences between cells derived from porcine aorta versus aortic valve Eur. J. Cardiothorac. Surg., October 1, 2003; 24(4): 527 - 534. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
