ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
George L. Zorn, III
Richard N. Pierson, III
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pfeiffer, S.
Right arrow Articles by Azimzadeh, A. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pfeiffer, S.
Right arrow Articles by Azimzadeh, A. M.
Related Collections
Right arrow Lung - transplantation

Ann Thorac Surg 2001;72:1681-1690
© 2001 The Society of Thoracic Surgeons

Original article: general thoracic

Role of anti-Gal{alpha}1,3Gal and anti–platelet antibodies in hyperacute rejection of pig lung by human blood

Steffen Pfeiffer, MDa, George L. Zorn, III, MDa, Sean Kelishadi, BSa, Rafael Oriol, MDb, Philippe Wolf, MDc, Richard N. Pierson, III, MDa, Agnes M. Azimzadeh, PhD*a

a Department of Cardiac and Thoracic Surgery, Vanderbilt University Medical Center and VA Medical Center, Nashville, Tennessee, USA
b Inserm U504, Cell Glycobiology and Signaling, Villejuif, France
c Department of Transplantation Surgery, Centre Hospital Universitaire, Strasbourg, France

* Address reprint requests to Dr Azimzadeh, Department of Cardiac and Thoracic Surgery, 2986 The Vanderbilt Clinic, Nashville, TN 37232-5734, USA
e-mail: agnes.azimzadeh{at}mcmail.vanderbilt.edu

Presented at the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 29–31, 2001.


   Abstract
Top
 Abstract
Introduction
 Material and methods
 Results
 Comment
 Acknowledgments
 Discussion
 References
 
Background. Previous work has shown that antibodies against porcine antigens are an important trigger of hyperacute lung rejection (HALR). The relative importance of Gal{alpha}1,3Gal epitopes and other antigens, such as those expressed on pig platelet membranes or lung itself, has not been defined. This study compares the efficiency of three anti–pig antibody depletion strategies, and their efficacy with regard to attenuation of HALR.

Methods. Plasma pooled from three human donors was adsorbed against Gal{alpha}1,3Gal disaccharide or porcine platelet extract (PPE), or passed through pig lung vasculature. Whole blood reconstituted using adsorbed plasma was then used to perfuse piglet lung, and results were compared with unmodified human blood.

Results. Depletion of lung-reactive anti–Gal{alpha}1-3Gal antibodies was most efficient with the {alpha}Gal column (99% ± 0.5% vs 87% to 93% ± 11% for PPE and 92% to 95% ± 8% for lung, p < 0.01 vs {alpha}Gal column). PPE column tended to be more efficient (77% to 84% ± 12%) in removing anti-PPE antibodies than pig lung (66% to 70% ± 14%) or the {alpha}Gal column (56% to 63% ± 16%, p < 0.05). Lung survival and function with each antibody depletion strategy was improved relative to unmodified controls (mean survival >=146 minutes vs 8 minutes for controls). Although {alpha}Gal and lung adsorption yielded more consistent lung protection (survival beyond 2 hours) than did PPE, no approach proved significantly superior. Complement C3a elaboration at 10 minutes was attenuated >80% by each adsorption strategy, an effect that was most pronounced in the lung adsorption group (95%, p < 0.01). Histamine elaboration was blunted significantly by PPE adsorption but not in other groups (p < 0.05). Platelet but not leukocyte sequestration was decreased with antibody depletion compared with the nondepleted group (44% to 50% vs 82%, p < 0.01).

Conclusions. Each antibody depletion strategy tested significantly prolongs lung xenograft survival and function compared with unmodified human blood, but none was sufficient to reliably prevent HALR. Depletion of antibodies against both {alpha}Gal and additional cell membrane antigens, or control of antibody-independent pathogenic pathways, may be necessary to consistently prevent HALR.


   Introduction
Top
 Abstract
 Introduction
Material and methods
 Results
 Comment
 Acknowledgments
 Discussion
 References
 
Discordant xenografts undergo hyperacute rejection (HAR) within minutes to hours after reperfusion with the recipient’s blood. According to the widely accepted paradigm, naturally occurring xenoreactive antibodies (XNA), mainly immunoglobulin M (IgM), bind to antigen expressed on donor endothelial cells, leading to complement activation and endothelial cell damage. Depletion of XNA has been proven effective to positively influence graft outcome in kidney and heart [1–4] and in lung xenotransplantation [5–7]. Thus, XNA are primarily responsible for the physiologic and pathological features of HAR. In the pig-to-human combination, the Gal{alpha}1, 3Gal epitope is the most important antigen [8, 9]. The importance of other non-{alpha}Gal antigens has not yet been clearly defined, and some results suggest that non-{alpha}Gal antigen targets expressed on cell membrane proteins and lipids or uniquely in specific organs may trigger biologically important facets of HAR [6].

Different techniques to deplete the recipient plasma from antibody have been developed. Nonspecifically adsorbing all immunoglobulins by binding to a protein A or other immunoaffinity column is efficient in preventing HAR in pig hearts perfused with human blood [10], but is not attractive for clinical use because it results in hypogammaglobulinemia and thus reduced immunity. Organ perfusion was introduced in the early stages of xenotransplantation, and has the theoretical advantage of removing antibodies against tissue-specific antigens. Efficiency of different organs to remove antibodies against {alpha}Gal antigens varies [11], and the efficacy of perfusion of a given organ for preventing HAR of a subsequent test organ is also variable [6], supporting the hypothesis that organ-specific antibodies may be important. However, this approach carries potential risks, in that complement pathway products and other pathogenic factors may be activated during organ perfusion, which might adversely influence survival of a subsequent graft.

Selective column adsorption offers a clinically more attractive option compared with organ perfusion because of its limited effect on plasma protein levels, complement, and coagulation [12]. It could also be reused in a peritransplant setting.

This study was designed to: (1) test the efficiency of three different techniques (Gal{alpha}1,3Gal column, porcine platelet extract column, pig lung preperfusion) to deplete naturally occurring human xenoreactive antibodies ({alpha}Gal, non-{alpha}Gal, lung specific and unspecific) from freshly collected human blood; and (2) to define the role of the different types of antibodies on the clinical and biochemical facets of hyperacute lung rejection. Pig platelets express {alpha}Gal epitopes [13] that are recognized by human natural anti–gal antibodies. They also express pig cell membrane antigens other than Gal, and thus might remove tissue-specific non-{alpha}Gal antibodies from the plasma [14]. This was the rationale behind the idea of using porcine platelet extracts to deplete XNA from human blood. This approach was compared with Gal column and adsorption against pig lung with regard to efficacy for removing {alpha}Gal and platelet antibodies.


   Material and methods
Top
 Abstract
 Introduction
 Material and methods
Results
 Comment
 Acknowledgments
 Discussion
 References
 
Pig lung perfusion
Farm piglets (3 to 8 kg) were used in a well-established in situ ex vivo model as described previously [15, 16]. Animals were preanaesthetized with Ketamine (10 mg/kg) (Ketaset; Fort Dodge Animal Health, Fort Dodge, IA) and Xylazine (1 mg/kg) (Rompun; Bayer Pharmaceuticals, Shawnee Mission, KS), and kept under inhalative anesthesia (Isoflurane 0.5% to 5%) throughout the surgical procedure. After tracheotomy, median sternotomy, and heparinization (500 units/kg), pigs were exsanguinated and abdominal viscera removed. Pulmonary artery and left atrium were cannulated with flanged stainless steel cannulae, and the ascending aorta and superior vena cava were ligated. Bronchial and collateral flow were recovered through cannulation of the abdominal aorta and inferior vena cava. Lungs were flushed with 150 mL of 5% human albumin at room temperature and the lungs attached to the perfusion apparatus. Lungs were kept inflated at end inspiration until ventilation and perfusion were initiated. Warm ischemic time was typically less than 15 minutes.

Lung perfusion was initiated via a roller pump (Renal Systems, Minneapolis, MN) at 100 mL/min/kg body weight. Flow rate of blood flow through the lungs, measured in the pulmonary vein outflow tubing (Cliniflow II electromagnetic blood flow meter, Carolina Medical Electronics Inc, King, NC), pulmonary artery, left atrial, and airway pressures were monitored continuously (Marquette Electronics Inc, Milwaukee, WI). Pulmonary vascular resistance (PVR) was calculated using the equation: PVR [mm Hg/mL/min] = (pulmonary artery pressure - left atrial pressure)/measured flow. A pop-off valve was installed and set at 50 cm H2O to avoid excessive pulmonary artery pressure during periods of high pulmonary vascular resistance. The lungs were ventilated (Harvard Apparatus, South Natick, MA) with a balanced air mix (21% O2, 5% CO2) 12 to 15 times per minute. Tidal volumes of 10 mL/kg were set initially and adjusted visually for adequate inflation while avoiding peak inspiratory pressure over 14 cm H2O. At designated time points, gas exchange was quantified using 95% O2 and 5% CO2 for 1 minute with immediate sampling from postlung effluent. PaO2, PaCO2, and pH were determined on an ABL30 blood gas analyzer (Radiometer, Copenhagen, Denmark). An arbitrary study endpoint of 4 hours was established, and experiments "surviving" for this interval were terminated after final sample collection. Lung failure was defined as loss of oxygenation across the lungs (< 10 mm Hg difference in PaO2 [FiO2 = 100%] - PaO2 [room air]; normally > 150 mm Hg), loss of transpulmonary blood flow (< 5 mL/kg/min), or development of gross tracheal edema prohibiting lung ventilation.

Perfusate preparation
Human blood (type O) was collected from volunteer donors at the Vanderbilt Clinical Research Center, in accordance with protocols approved by the Committee for the Protection of Human subjects. Approximately 400 mL of fresh human blood from one donor was collected in blood collection bags (Boin Medica Corporation, Seoul, Korea) containing 100 mL citrate phosphate dextrose adenine solution (CPDA). Two units of pooled fresh-frozen plasma from matching blood type was added to reach a baseline perfusion volume of 1,000 mL. For antibody depletion experiments, the blood was separated into its cellular and plasma components by centrifugation, and the supernatant plasma was then, together with the pooled plasma, depleted from XNA using one of the three methods described below; plasma was kept cold until reconstituted. The perfusate was then heparinized with a plasma concentration of 2 IU/mL of heparin (Elkins-Sinn, Cherry Hill, NJ), recalcified with 1.3 mg/mL of calcium chloride (to restore normal ionized calcium levels in CPDA banked blood products; Fujisawa, Deerfield, IL) and the pH adjusted to physiologic values using sodium bicarbonate 8.4% (American Pharmaceutical Partners, Los Angeles, CA) based on blood gas results. The perfusate was circulated at 37°C for approximately 15 minutes before baseline blood samples were collected and perfusion of the lung was started.

XNA antibody depletion
After blood collection, plasma and cellular blood components were separated by centrifugation (Sorvall RT 6000B, 3200 rpm, 15 minutes, 10°C; Du Pont, Newton, CT). The freshly collected plasma and thawed fresh-frozen plasma were combined and adsorbed. Plasma samples were taken before and after the depletion process to measure efficiency of antibody removal and for assays of complement activation products.

Alpha Gal column
{alpha}Gal antigen (Gal{alpha}1-3Galß1) bound to sepharose 6FF matrix beads via polyacrylamide (PAA-Bdi) was a kind gift from Syntesome (Moscow, Russia). The beads were kept in a column measuring 5 cm in diameter and 8 cm in height (total volume of gel was 200 mL). Before use, the columns were flushed with 11 sterile saline. Plasma was passed through the column at room temperature at a flow rate of 25 mL/min to allow enough time for adsorption. The column was then regenerated by eluting bound antibodies with 0.1 mol/L glycine buffer (pH 2.5) and stored at 4°C in phosphate-buffered saline (PBS) containing 0.05% NaN3 until next use.

Porcine platelet extract (PPE) column
Porcine platelet extracts were obtained and bound to sepharose 4B according to the technique described elsewhere [17]. The beads were kept in two columns (both used for each experiment), each measuring 5 cm in diameter and 8 cm in height (total volume of gel was 350 mL). Perfusion, antibody elution, and storage technique were the same as with the {alpha}Gal column.

Pig lung perfusion
Pig lungs were isolated according to the above-described protocol. Plasma was passed once through the lung and collected. To prevent vasoconstriction of the lung, epoprostenol (Flolan; Burroughs Welcome Co, London, England) was administered to the plasma (0.5 mg) and injected into the pig as an IV bolus (1.0 mg) before exsanguination.

Histology and immunofluorescence
Lung biopsies were obtained preperfusion, at 10 minutes, and at lung failure or after 4 hours. The lung tissue was gently infused with Tissue-Tek OCT (diluted 1:1 in phosphate-buffered saline), snap frozen in liquid nitrogen, and stored at -70°C. Six-micron sections were prepared, fixed in acetone, and subsequently stained for immunohistological analysis as described previously [18]. Monoclonal antibodies anti-IgG (Pharmingen, San Diego, CA), C3 (Dako, Copenhagen, Denmark), anti-IgM (Immunotech, Westbrook, ME), and C1q (Quidel, San Diego, CA) (diluted at 1:50, 1:5, 1:50, and 1:100, respectively) were used to assess XNA and complement deposition in the graft. The reactivity of monoclonal antibodies was revealed using a Cy3-labeled goat anti–mouse IgG (H + L) (Jackson Laboratories, West Grove, PA). For identification of deposits on endothelium, double-fluorescence staining was used on each section using a rabbit anti–von Willebrand factor antibody (Dako), revealed by a FITC-labeled goat anti–rabbit IgG (H+L) antibody (Jackson Labs). Lung tissue from an autologous experiment (pig lung perfused with pig blood) was used as a negative control in all staining experiments. Sections were examined using a Zeiss Axioplan 2 two-color fluorescent microscope equipped with Zeiss Image 3.0 (Carl Zeiss Inc, Thornwood, NY) image acquisition software.

Histamine, C3a, and sC5b-9 assays
Plasma samples stored in EDTA at -70°C were assayed by commercial enzyme-linked immunosorbent assay (ELISA) for histamine (Immunotech) and C3a and sC5b-9 (Quidel), according to the manufacturers’ instructions.

Blood cell analysis
Blood samples were collected in EDTA during the experiment, and white blood cells, red blood cells, and platelets were measured by standard hematology techniques.

ELISA for measurement of anti–pig antibody
Anti–{alpha}Gal antibody
Determination of anti–{alpha}Gal antibodies was performed by ELISA as described [19]. In brief, Maxisorp microtiter plates (Nalgene Nunc Int, Rochester, NY) were coated with 5 µg/mL Gal{alpha}1-3Gal-polyacrylamide conjugate (PAA-Bdi) (GlycoTech Corporation, Rockville, MD) overnight at 4°C. After washing, serum samples were incubated for 90 minutes at 37°C in PBS containing 1% BSA and 0.05% Tween 20. The binding of IgG and IgM binding was revealed by addition of affinity-purified alkaline phosphatase (AP)-conjugated goat anti–human Fc{gamma} and Fcµ antibodies (Jackson Laboratories, West Grove, PA) diluted in PBS-T-BSA. Results were expressed in arbitrary units, calculated from human serum used as an internal standard, and also expressed as the percentage of antibody adsorption.

Anti–PPE antibody
Levels of anti–PPE antibodies were measured by ELISA using porcine platelet extracts (PPE) as antigen. Platelets were isolated from a pool of 10 large white pigs, and lysates were prepared as previously described [17] and stored at -70°C until use. Maxisorp microtiter plates (Nunc) were coated overnight at 4°C with PPE extracts (12 µg/mL) in Tris-buffered saline (TBS). After rinsing with TBS containing 0.05% Tween 20 (TBS-T), 1% BSA in TBS-T was added for 1 hour at 37°C. Serum samples standard serially diluted in TBS-T-BSA were added for 2.5 hours at room temperature. The binding of IgG and IgM was revealed by addition of AP-conjugated goat anti–human Fc{gamma} and Fcµ antibodies (Jackson Labs) for 1 hour at room temperature, and absorbance was read at 405 nm on a Dynatech MRX plate reader (Dynatech Laboratories, Chantilly, VA). Results were expressed in arbitrary units, calculated from human serum used as an internal standard, and also expressed as the percentage of antibody adsorption.

Statistical analysis
Data are presented as mean ± standard deviation for all variables. Continuous variables were checked for normality by plotting histograms. Variables that were not normally distributed were analyzed using the Kruskal-Wallis and the Mann-Whitney U tests. Those that were normally distributed were assessed with a one-way analysis of variance and the Student’s t test. Correlation between either survival or PVR and other parameters inside each group was studied using the Pearson correlation test. Values of p less than 0.05 were considered statistically significant. All tests were two tailed. All statistical analyses were performed on a personal computer with the statistical package SPSS for Windows (version 10.0).

Animal care and use
All experiments were conducted under protocols approved by the Vanderbilt Animal Care and Use Committee and in accordance with the guidelines from National Institutes of Health publication 86-23, "Guide for Care and Use of Laboratory Animals."


   Results
Top
 Abstract
 Introduction
 Material and methods
 Results
Comment
 Acknowledgments
 Discussion
 References
 
Lung survival and function
Pig lungs perfused with autologous blood (n = 3) consistently survived for over 4 hours with a peak PVR lower than 0.05 mm Hg/mL/min throughout the experiment. Xenogeneic grafts (n = 7) were hyperacutely rejected after 8.3 ± 3.3 minutes, with none surviving 20 minutes (Fig 1). Five fulfilled the flow criteria due to excessively high (PVRmax = 0.62 ± 0.33 mm Hg/mL/min at 5 minutes); two grafts were lost due to early tracheal edema.



View larger version (13K):
[in this window]
[in a new window]
  		Fig 1. Kaplan-Meier cumulative survival (%) of ex vivo perfused lungs. Survival of lungs from {alpha}Gal and lung preperfusion groups are almost identical and therefore superimposed. (PPE = porcine platelet extract.)

 
Antibody depletion by each method increased survival time of the lung (p < 0.01 vs unmodified human blood). Although substantial sequestration of blood perfusate was grossly evident in every experiment, none of the adsorption treatment group grafts failed by meeting loss of oxygenation criteria. All grafts reaching the perfusion time of 4 hours oxygenated well until the end of the study.

Gal-group lungs (n = 5) survived for 216 ± 39 minutes, three organs reaching the study endpoint of 240 minutes reperfusion. The two organ losses were due to sequestration of most of the blood perfusate into the lung parenchyma. Peak PVR resistance was significantly lower (0.10 ± 0.09 at 10 minutes, p < 0.01) compared with unmodified blood controls.

In the PPE-group (n = 5), three grafts were lost early at 15, 80, and 155 minutes. Two grafts accumulated interstitial fluid into the lung parenchyma; one graft was lost to rise in PVR with loss of flow through the graft. Mean survival with PPE adsorption was 146 ± 99 minutes. Peak PVR was 0.35 ± 0.49 at 10 minutes, and due to the one graft loss because of high PVR, it was not significantly different from the xeno-control group. Two organs reached the endpoint with good pulmonary function throughout the experiment.

One graft from the lung group (n = 5) was lost to tracheal edema, and one to sequestration of blood volume into the graft. Mean survival after pig lung adsorption was 219 ± 33 minutes. Peak PVR (15 minutes) was 0.11 ± 0.10, and significantly lower than peak PVR in the unmodified blood control group (p = 0.01).

Perfusate XNA levels
With the exception of anti-{alpha}Gal IgM levels in the lung perfusion group, total IgM and IgG XNA against {alpha}Gal and PPE measured before any intervention were statistically similar between groups. The anti–{alpha}Gal antibody levels in pooled human plasma were 147 ± 70 arbitrary units (IgM) and 89 ± 77 arbitrary units (IgG); and anti-PPE levels were 124 ± 78 arbitrary units (IgM) and 138 ± 129 arbitrary units (IgG), respectively.

Antibody depletion
Total antigen-specific anti–{alpha}Gal and anti–PPE antibodies (IgM and IgG) were reduced by each adsorption procedure, and are summarized in Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Antigen-Specific Antibody Depletion

 
The {alpha}Gal column depleted 97% ± 2% of anti–{alpha}Gal IgM and 96% ± 3% of anti–{alpha}Gal IgG, being more efficient than the two other treatment strategies (p < 0.01 for IgG and IgM vs PPE and lung perfusion groups). Best depletion of anti-PPE antibodies was achieved with the PPE column, leading to reduction of total PPE antibodies to 15% ± 8% and 23% ± 14% for IgM and IgG, respectively. Pig lungs removed antigen-specific anti–PPE IgM antibodies better than the {alpha}Gal column (Table 1).

The total level of lung-specific antibodies detected in each assay was calculated by subtracting the antibody titer (arbitrary units) measured after 30 minutes of lung perfusion from the antibody titer (arbitrary units) at the beginning of the perfusion. The percent of lung-specific antibody that was reactive with {alpha}Gal or PPE, as measured in {alpha}Gal or PPE ELISA, and that was removed by each adsorption procedure, was then calculated and expressed as a percentage of antigen-specific antibodies.

The {alpha}Gal column removed all lung-specific anti–{alpha}Gal IgM and IgG (> 99%), being more efficient than the two other depletion strategies (p < 0.01; Table 2, Fig 2). Lung-specific anti–PPE antibodies were best removed by pig lungs and PPE column, and results were significantly superior to removal by the {alpha}Gal column (Table 2, Fig 3).


View this table:
[in this window]
[in a new window]
  		Table 2. Lung-Specific Antibodies

 


View larger version (10K):
[in this window]
[in a new window]
  		Fig 2. Anti–{alpha}Gal antibody levels in arbitrary units (AU) over time. Different antibody class (IgM, IgG) levels measured in different treatment groups. Efficiency of the different treatment strategies for depletion of anti–{alpha}Gal directed antibodies (IgM and IgG) (expressed as % remaining = measurable antibody). Values corrected for lung specificity by subtracting levels of antibody detectable at 30 minutes of lung perfusion experiment (and therefore not binding to lung tissue) from baseline values using the equation: % remaining antibody = Titer0min - Titer30 min/Titerpre absorption - Titer30 min x 100%. Figures of more detailed illustration of efficiency during single experiments are available directly from the authors. (PPE = porcine platelet extract.)

 


View larger version (14K):
[in this window]
[in a new window]
  		Fig 3. Anti–PPE antibody levels in arbitrary units (AU) over time. Different antibody class (IgM, IgG) levels measured in different treatment groups. Efficiency of the different treatment strategies for depletion of anti–PPE directed antibodies (IgM and IgG) (expressed as % remaining = measurable antibody). Values corrected for lung specificity by subtracting levels of antibody detectable at 30 minutes after lung perfusion (and therefore not binding to lung tissue) from baseline values using the equation: % remaining antibody = Titer0 min - Titerpre absorption - Titer30 min x 100%. Figures of more detailed illustration of efficiency during single experiments are available directly from the authors. (PPE = porcine platelet extract.)

 
Complement activation assays
Levels of C3a and sC5b-9 increased throughout the experiments in all groups (Figs 4, 5). The xeno controls showed the highest rise of C3a (243 ± 57 ng/mL) and of sC5b-9 (137 ± 25 ng/mL) within 1 minute. All methods of Ab depletion controlled complement activation, producing significantly less C3a and sC5b-9 at 1 and 10 minutes compared with controls (p < 0.05, except sC5b-9 at 10 minutes in the PPE group). Complement activation was best controlled in the lung perfusion group, which produced significantly less C3a at 10 minutes (118 ± 56 ng/mL) and at 60 minutes (322 ± 122 ng/mL) than the PPE group (566 ± 291 and 1,259 ± 344 ng/mL, respectively). In depletion groups, complement activation products C3a and sC5b-9 tended to be higher in the PPE group.



View larger version (16K):
[in this window]
[in a new window]
  		Fig 4. Elaboration of complement C3a during the experiment expressed (in ng/mL) as levels at the respective timepoint - levels at the beginning of the experiment (t = 0 minutes).

 


View larger version (15K):
[in this window]
[in a new window]
  		Fig 5. Elaboration of soluble complement C5b-9 during the experiment expressed (in ng/ml) as levels at the respective timepoint - levels at the beginning of lung perfusion (t = 0 minutes).

 
Histamine
Production of histamine at 10 minutes was lowest in the PPE group (36 ± 30 ng/mL) and significantly lower (p < 0.05) than in the unmodified human blood control group (103 ± 57 ng/mL). Other treatment strategies were not statistically different from unmodified human blood controls.

Extravasation of cellular blood components
The hematocrit at the beginning of the experiments was 15 ± 2 mg/dL for all groups and was similar in the different groups. Hematocrit was stable throughout the experiments, suggesting no preferential sequestration of red blood cells or plasma in the lung parenchyma. No evidence of gross hemolysis was appreciated in plasma supernatants.

In unmodified blood controls, 83% ± 12% of the platelets sequestered into the lung parenchyma within 1 minute. All antibody depletion strategies preserved platelets at 50% ± 10% of baseline values (p < 0.05 vs unmodified blood controls). No difference was appreciated between depletion groups with regard to inhibition of platelet sequestration in the lung.

In the unmodified human blood group, 64% ± 13% of leukocytes sequestered into the lung parenchyma within 10 minutes. Antibody depletion strategies did not prevent white blood cell sequestration; cell counts were similarly reduced by 65% ± 21%.

Immunohistology
Immunoglobulin deposits, strong C1q and C3b (++ to +++), and mild C5b-9 (+) deposition were found in vessels and capillaries of all lungs perfused with unmodified human blood. C3b and C1q deposition was diminished (+) or absent (-) in all treatment group lungs, with no difference between groups. PPE group grafts consistently showed deposition of human IgG (+ to +++), but not IgM (- to +) (not shown). A mild deposition of IgG and IgM was evidenced in lungs from the Gal group. Weak C3b, but not C1q, deposition could be detected in the pig lung perfusion group (Fig 6). Antibody deposits did not correlate with graft survival.



View larger version (102K):
[in this window]
[in a new window]
  		Fig 6. Immunohistological analysis of pig lungs after 10 minutes of perfusion with human blood (staining for IgM in the left panels, for C3b in the right panels). Lungs used for antibody adsorption show bright staining for IgM and C3b. Lungs from the {alpha}Gal group show week staining for IgM and IgG, suggesting that despite depleting all {alpha}Gal-specific antibodies, there are still lung-specific antibodies binding to the graft during perfusion. Staining for C3b is negative, indicating that those bound antibodies do not activate complement. Lungs from lung perfusion group do stain negative for antibodies, suggesting that all lung-sensitive antibodies had been depleted.

 

   Comment
Top
 Abstract
 Introduction
 Material and methods
 Results
 Comment
Acknowledgments
 Discussion
 References
 
We and others have previously demonstrated an important role for xenoreactive antibodies in hyperacute lung rejection [5–7]. Our current findings document the relative efficiency of three strategies to deplete xenoreactive anti–pig antibody from human plasma. We find that column immunoadsorption removes the majority of antigen-specific antibody, but small amounts of lung-reactive antibody of that specificity are residual after adsorption as employed.

Similarly, some antigen-specific antibody remains after preperfusion using a single passage through porcine lung. For example, in three of the five experiments, up to 16% of the detectable Gal-specific antibody does not bind to either the preperfusion lung or the test lung, suggesting that some anti–{alpha}Gal antibodies (mainly IgM) recognize Gal epitopes, which are not expressed in the lung. More important from a practical perspective, in two of the five experiments, the degree of lung-specific antibody depletion was insufficient to prevent IgM binding to the graft, which may be presumed to contribute to antibody-mediated injury. Variation between experiments may reflect the interindividual variability of the expression of {alpha}Gal epitopes between individual pigs, and in anti–Gal antibody repertoire between individual humans. Whether use of a second lung to further deplete lung-specific antibodies would result in improved protection of a subsequent test graft was not addressed in this study. However, more efficient antibody depletion using a combination of strategies or more efficient columns should result in improved survival and function.

Alternatively, the lung may be particularly sensitive to antibody-independent mechanisms of injury. This mechanistic distinction has important practical consequences for the development of lung xenografting, as highly efficient, prolonged antibody absorption will be difficult to achieve in vivo, and rational development of alternative or additional approaches depends on the result. Further investigations will be necessary to determine whether the physiologic and biochemical aspects of HALR observed after antibody adsorption are attributable to residual anti–Gal or anti–PPE antibody, or directed against other antigens expressed in the lung.

Survival of lungs was significantly prolonged in association with each antibody adsorption strategy, confirming that the antibodies removed by each approach are important to the pace and character of HALR. However, no single strategy completely abrogated the physiologic and biochemical perturbations characteristic of HALR.

Two previous studies have proposed that lung-specific antigens may account for the particular vulnerability of this organ to HALR. Macchiarini and associates [6] reported from an ex vivo lung perfusion model that lungs perfused with human blood depleted from antibodies by pig lung preperfusion performed better compared with lungs perfused with blood depleted from antibodies by liver, spleen, or {alpha}Gal column preperfusion. It was shown that IgM antibodies specific for low-molecular weight nongalactosylated proteins were removed only by pig lung perfusion, suggesting a functional role for lung-specific non-{alpha}Gal antibodies in HALR. Similarly, using pig lung donors transgenic for hDAF and CD59 in an in vivo primate lung transplant model, Lau and associates [7] found improvement of graft outcome by preperfusion of the recipient blood through a pig lung, but not by using a pig kidney or after total immunoglobulin depletion using Therasorb columns. In contrast, our study does not demonstrate an advantage to using lung relative to Gal antigen-specific or PPE-reactive factors. We speculate that the difference may be explained by the fact that both Macchiarini and Lau and associates passed whole blood through the lung. As demonstrated by Daggett and associates [20], by our previous work [15], and current observations, the lung also reduces neutrophils by 85% and platelets by 95%. From our own observations, we know that platelets (Pfeiffer and associates, manuscript in preparation) and neutrophils [16] also contribute to complement-mediated lung injury during HAR. Indeed, inhibiting platelet adhesion and aggregation blunts complement elaboration during pig lung perfusion with human blood (our own observations, presented at the 20th ISHLT meeting, Osaka, 2000). We infer that the relative benefit of lung adsorption in these studies is consequent to the particular avidity of the lung for pathogenic human cellular elements, and not solely a consequence of particular efficiency in antibody depletion, or direct evidence for existence of lung-specific antibodies. This hypothesis is also consistent with the fact that long-term graft survival has proven remarkably difficult to achieve in any of several life-supporting lung xenotransplantation models (our own unpublished observations) [20, 21].

Each strategy we tested has potential advantages and disadvantages. Organ depletion methods offer an efficient way of removing xenoreactive antibodies, but they generate complement and coagulation activation, which may be deleterious to the graft [22]. The efficiency of antibody depletion using columns to prevent HALR has been limited so far. Pig platelet glycoproteins are recognized by IgM and IgG {alpha}Gal antibodies [13], and share several common determinants with pig endothelial cell glycoproteins [14]. An immunoaffinity column bearing pig platelet proteins was produced, and its efficacy to remove anti–Gal and anti–pig endothelial cell antibodies was assessed in vitro and in vivo (manuscript in preparation) [23]. In this study, we tested whether an affinity column bearing pig platelet proteins would deplete lung-specific anti–Gal and non-Gal antibodies, and thereby, could replace the efficient but crude lung perfusion method. We also wanted to evaluate the specific role of antibodies in HALR, by comparing columns and lung depletion using plasma, and not whole blood.

All methods of antibody depletion improved graft survival dramatically but not reliably. In each of the treatment groups, grafts were lost early before reaching the study endpoint of 4 hours reperfusion. We could not find a correlation between specific antibody (anti {alpha}Gal or PPE) levels at the beginning of an experiment and survival time comparing the different treatment groups. Each approach to antibody depletion was efficient to deplete >90% of lung-specific antibody, and none yielded significantly superior protection of the lung. Either 90% depletion is not sufficient, or antibody independent mechanisms may be critically important to HALR.

According to the recognized paradigm, binding of antibodies, mainly {alpha}Gal IgM, activates complement and is one of the main contributors to HALR. Finding a correlation between anti–{alpha}Gal IgM titers and complement elaboration at 10 minutes (all groups combined) but no other measured antibody levels at the beginning of the experiment (p < 0.05) supports these findings. But, despite more effective {alpha}Gal (IgG and IgM) depletion with the {alpha}Gal column, complement elaboration was not better controlled compared with other groups. This suggests that depleting {alpha}Gal antibodies is as efficient as the pig lungs to prevent complement activation.

In conclusion, the injury to lung pig-to-primate transplants during HAR is very complex and seems to differ from other organs. Xenoreactive naturally occurring antibodies play a major role, but depleting them with the means that we have to date, and that are already used to prevent HAR of other organs, is not sufficient to completely prevent HALR. Additional strategies like inhibiting complement elaboration and activation of platelets may be necessary. The development of lung-specific antibody depletion strategies such as combining {alpha}Gal and pig lung perfusion to deplete antibodies may help to control HALR better and more reliably.


   Acknowledgments
Top
 Abstract
 Introduction
 Material and methods
 Results
 Comment
 Acknowledgments
Discussion
 References
 
The authors want to thank Daniel Byrne for his expertise in statistical analysis and Nicolai Bovin from Syntesome for the kind gift of the Gal{alpha}1,3Gal immunoabsorbent. We also want to thank Lynn Giovanoni and Leslie Sisemore for their expert technical assistance during performance of the experiments and in vitro analysis.


   Discussion
Top
 Abstract
 Introduction
 Material and methods
 Results
 Comment
 Acknowledgments
 Discussion
References
 
DR JOSEPH B. ZWISCHENBERGER (Galveston, TX): Just to clarify, your definition of survival is that your isolated pumped lung prep maintains perfusion for 4 hours?

DR PFEIFFER: Correct. Since we only wanted to study events during hyperacute rejection, which happens in our model within 10 minutes in untreated controls, we did not see a reason to go longer.

DR ZWISCHENBERGER: Your {alpha}Gal and your pig lung filtration both seem equivalent, but they clearly have different mechanisms.

DR PFEIFFER: This is how it seems to be. In summary, if we deplete {alpha}Gal antibodies, we still find IgM and IgG binding to the lungs during perfusion, suggesting that other, lung-specific non-{alpha} Gal antibodies may be responsible for the damage. On the other hand, if we deplete antibodies binding with lung perfusion, we do not find those antibodies binding to the graft during perfusion, but the grafts are still rejected. This suggests other than antibody-related mechanisms contributing to HALR.

DR WILLIAM H. WARREN (Chicago, IL): To what degree can these findings be extrapolated to xenotransplantation of pig hearts to human?

DR PFEIFFER: As I am sure you know, antibody depletion has been used in heart and kidney transplantation, both to control HAR, and also to prevent and treat delayed humoral rejection.

The lung seems to be a special case, meaning that the strategies that work in other organs to prevent hyperacute rejection do not seem to be enough for the lungs. By simply controlling the formation of membrane attack complex through complement-directed therapy, either using transgenic pigs or soluble complement inhibitors, HAR can be prevented for kidney and heart. Other groups and we were able to show that this does not apply to the lungs.

If you are referring to the possible importance of organ-specific antigens and antibodies, yes, I can image that the same applies to hearts, meaning that there may be different heart- and lung-specific antibodies. But it would be technically challenging to address this question directly. For example, to efficiently deplete antibodies by heart perfusion may require serial plasma perfusion through several hearts, because I could imagine that heart perfusion would not be as efficient considering the size of the capillary bed of a heart, which is very small compared to the lung.

Whether adsorption against lung works to treat delayed humoral rejection in the lungs, we do not know, since we are not that far yet. We have not been able to achieve in vivo long-term survival in lung xenotransplantation of sufficient duration to examine this problem.


   References
Top
 Abstract
 Introduction
 Material and methods
 Results
 Comment
 Acknowledgments
 Discussion
 References
 

  1. Kozlowski T., Ierino F.L., Lambrigts D., et al. Depletion of anti-Gal({alpha})1-3Gal antibody in baboons by specific {alpha}-Gal immunoaffinity columns. Xenotransplantation 1998;5:122-131.[Medline]
  2. Kozlowski T., Shimizu A., Lambrigts D., et al. Porcine kidney and heart transplantation in baboons undergoing a tolerance induction regimen and antibody adsorption. Transplantation 1999;67:18-30.[Medline]
  3. Taniguchi S., Neethling F.A., Korchagina E.Y., et al. In vivo immunoadsorption of antipig antibodies in baboons using a specific Gal({alpha})1-3Gal column. Transplantation 1996;62:1379-1384.[Medline]
  4. Brenner P., Reichenspurner H., Schmoeckel M., et al. IG-therasorb immunoapheresis in orthotopic xenotransplantation of baboons with landrace pig hearts. Transplantation 2000;69:208-214.[Medline]
  5. Pierson R.N., III, Kasper-Konig W., Tew D.N., et al. Hyperacute lung rejection in a pig-to-human transplant model: the role of anti-pig antibody and complement. Transplantation 1997;63:594-603.[Medline]
  6. Macchiarini P., Oriol R., Azimzadeh A., et al. Evidence of human non-{alpha}-galactosyl antibodies involved in the hyperacute rejection of pig lungs and their removal by pig organ perfusion. J Thorac Cardiovasc Surg 1998;116:831-843.[Abstract/Free Full Text]
  7. Lau C.L., Daggett W.C., Yeatman M.F., et al. The role of antibodies in dysfunction of pig-to-baboon pulmonary transplants. J Thorac Cardiovasc Surg 2000;120:29-38.[Abstract/Free Full Text]
  8. Holzknecht Z.E., Coombes S., Blocher B.A., Lau C.L., Davis R.D., Platt J.L. Identification of antigens on porcine pulmonary microvascular endothelial cells recognized by human xenoreactive natural antibodies. Lab Invest 1999;79:763-773.[Medline]
  9. Joziasse D.H., Oriol R. Xenotransplantation: the importance of the Gal{alpha}1,3Gal epitope in hyperacute vascular rejection. Biochim Biophys Acta 1999;1455:403-418.[Medline]
  10. Brenner P., Hinz M., Huber H., et al. The influence of antibody and complement removal with an Ig-Therasorb column in a xenogeneic working heart model. Eur J Cardiothorac Surg 1999;15:672-679.[Abstract/Free Full Text]
  11. Azimzadeh A., Meyer C., Watier H., et al. Removal of primate xenoreactive natural antibodies by extracorporeal perfusion of pig kidneys and livers. Transpl Immunol 1998;6:13-22.[Medline]
  12. Meyer C., Wolf P., Romain N., et al. Use of von Willebrand diseased kidney as donor in a pig-to-primate model of xenotransplantation. Transplantation 1999;67:38-45.[Medline]
  13. Thibaudeau K., Borche L., Soulillou J.P., Blanchard D. Characterization of porcine platelet glycoproteins recognized by human natural "anti-gal" antibodies. Blood 1996;87:4636-4642.[Abstract/Free Full Text]
  14. Platt J.L., Lindman B.J., Chen H., Spitalnik S.L., Bach F.H. Endothelial cell antigens recognized by xenoreactive human natural antibodies. Transplantation 1990;50:817-822.[Medline]
  15. Blum M.G., Collins B.J., Chang A.C., Zhang J.P., Knaus S.A., Pierson R.N., III Complement inhibition by FUT-175 and K76-COOH in a pig-to-human lung xenotransplant model. Xenotransplantation 1998;5:35-43.[Medline]
  16. Collins B.J., Blum M.G., Parker R.E., et al. Thromboxane mediates pulmonary hypertension and contributes to lung inflammation during hyperacute lung rejection. J Applied Physiol 2001;90:2257-2268.[Abstract/Free Full Text]
  17. Azimzadeh A., Romain N., Vermot-Desroches C., et al. Antibodies to human adhesion molecules and von Willebrand factor: in vitro cross-species reactivity in the xenotransplantation setting. Xenotransplantation 1998;5:284-291.[Medline]
  18. Zhang J.P., Blum M.G., Chang A.C., et al. Immunohistologic evaluation of mechanisms mediating hyperacute lung rejection, and the effect of treatment with K76-COOH, FUT-175, and anti-Gal column immunoadsorption. Xenotransplantation 1999;6:249-261.[Medline]
  19. Rieben R., Von Allmen E., Korchagina E.Y., et al. Detection, immunoabsorption, and inhibition of cytotoxic activity of anti-aGal antibodies using newly developed substances with synthetic Gal a1-3Gal disaccharide epitopes. Xeno 1995;2:98-106.
  20. Daggett C.W., Yeatman M., Lodge A.J., et al. Total respiratory support from swine lungs in primate recipients. J Thorac Cardiovasc Surg 1998;115:19-27.[Abstract/Free Full Text]
  21. Blum MG, Chang AC, Collins BJ, et al. The effect of nitric oxide and thromboxane blockade on pulmonary vascular resistance in a pig to primate lung transplant model [Abstract] The 4th International Congress for Xenotransplantation September 7–11, 1997.
  22. Meyer C., Wolf P., Romain N., et al. Use of von Willebrand diseased kidney as donor in a pig-to-primate model of xenotransplantation. Transplantation 1999;67:38-45.
  23. Azimzadeh A, Kaufmann F, Lucchiari N, et al. In vitro and in vivo immunoadsorption of anti-pig antibodies using a porcine platelet protein column [Abstract]. The 4th International Congress for Xenotransplantation September 7–11, 1997.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
George L. Zorn, III
Richard N. Pierson, III
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pfeiffer, S.
Right arrow Articles by Azimzadeh, A. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pfeiffer, S.
Right arrow Articles by Azimzadeh, A. M.
Related Collections
Right arrow Lung - transplantation

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