Flow Cytometric Measurement of ABO Antibodies in ABO-Incompatible Living Donor Kidney Transplantation Gisella Puga Yung,1 Piero V. Valli,1 Astrid Starke,2 Regula J. Mueller,1 Thomas Fehr,2 ¨ zpamir,3 Urs Schanz,3 Markus Weber,4 Rudolf P. Wu¨thrich,2 Marija Cesar-O Jo¨rg D. Seebach,1,5 and Georg Stussi1 Due to different detection methods, a comparison of anti-A/B antibody (Ab) levels among transplantation centers after living donor ABO-incompatible kidney transplantation is problematic. In the present study, anti-A/B Ab levels were determined prior to, and after, blood group A-to-O kidney transplantation using a recently established semiquantitative flow cytometry-based method, ABO fluorescence-activated cell sorting (ABO-FACS), and compared with standard agglutination titers and indirect antiglobulin testing. Pretransplant agglutination titers were reduced from 1:64 to 1:4, by a total of 14 Glycosorb A column immunoadsorptions (IADSs). Compared with the agglutination titers, antidonor immunoglobulin (Ig) M ABO-FACS mean fluorescence intensity ratios (MFIRs) decreased faster and remained low. No difference was observed using donor type or third-party A red blood cells (RBCs) for the ABO-FACS. Glycosorb A columns were not specific, also reducing anti-B and antiporcine IgM levels, which was confirmed by detecting anti-A/B and antiporcine Abs in the column eluates. In conclusion, analysis of pre- and posttransplant Abs from ABO-incompatible kidney transplant recipients by ABO-FACS allows a better understanding of Ab kinetics, which may improve the design of future IADS protocols. Keywords: ABO fluorescence-activated cell sorting (ABO-FACS), ABO-incompatible living donor kidney transplantation. (Transplantation 2007;84: S20–S23)
ver the past few years, ABO-incompatible living donor kidney transplantation has become a valuable approach to enlarging the available donor pool (1– 6). According to Landsteiner’s seminal discovery, individuals with blood group A produce anti-B antibodies (Ab), while individuals with blood group B produce anti-A Ab, and individuals with blood group O produce both anti-A and anti-B Ab. Since ABO antigen (Ag) is expressed not only on red blood cells (RBCs), but also on endothelial cells (ECs) (7, 8), anti-A/B Ab can bind to ECs of the donor organ and induce (hyper-) acute humoral graft rejection. Therefore, circulating anti-A/B Abs need to be reduced in the recipient’s serum prior to transplantation in order to overcome anti-A/B Ab-mediated rejection and early graft loss in ABO-incompatible kidney transplantation (5, 9, 10). To achieve this goal, several techniques have been reported, including Ag-specific immunoadsorption (IADS), plasma exchange, and double filtration (1, 4, 5, 11). Similarly
O
The authors declare no potential conflicts of interest. G. Puga Yung and G. Stussi contributed equally to this manuscript. Supported by grants from the Swiss National Science Foundation (404658668), the University of Zurich (FK 54170101), Union Banque Swiss AG on behalf of a client, and the Krebsliga Zu¨rich. 1 Laboratory for Transplantation Immunology, University Hospital Zurich, Zurich, Switzerland. 2 Clinic of Nephrology, University Hospital Zurich, Zurich, Switzerland. 3 Clinic of Hematology, Department of Internal Medicine, University Hospital Zurich, Zurich, Switzerland. 4 Department of Visceral and Transplant Surgery, University Hospital Zurich, Zurich, Switzerland. 5 Address correspondence to: Jo¨rg D. Seebach, M.D., University Hospital Zurich, Department of Internal Medicine, Laboratory for Transplantation Immunology, Ra¨mistrasse 100, C HOER 31, CH-8091 Zurich, Switzerland. E-mail:
[email protected] Copyright © 2007 by Lippincott Williams & Wilkins ISSN 0041-1337/07/84012S-20 DOI: 10.1097/01.tp.0000296646.17845.12
S20
to the ABO Ag, the carbohydrate Gal␣1-3Gal (␣Gal) Ag, which is ubiquitously expressed in vertebrates except in humans and anthropoid apes, stimulates the generation of anti␣Gal Ab in humans (8, 12). These anti-␣Gal Abs bind to porcine cells, causing hyperacute rejection in xenotransplantation (13), and represent an excellent positive control to evaluate anti-A/B Ab. Standard immunohematologic evaluation of anti-A/B Ab titers includes blood group typing, agglutination, and indirect antiglobulin testing (IAT), providing only indirect evidence for the presence of immunoglobulin (Ig) G Abs and being unable to detect IgG subclasses. The major disadvantage of these assays is the lack of standardized protocols, with considerable interobserver and interassay variability, as well as the failure to clearly discriminate IgG from IgM Abs. The former leads to a poor comparability of the titers originating from different centers. The latter contributes to the confusion about the role of IgG (subclasses) versus IgM Abs in humoral rejection. To resolve these problems, we utilized a recently established semiquantitative flow cytometry-based method, ABO fluorescence-activated cell sorting (ABO-FACS), to measure anti-A/B and antiporcine IgM/IgG Ab serum levels prior to, and after, a living-donor ABO-incompatible kidney transplantation (14, 15).
MATERIALS AND METHODS ABO Fluorescence-Activated Cell Sorting The ABO-FACS method has been described in detail elsewhere (14). In brief, rhesus-negative A or B RBCs, donor RBCs, and porcine RBCs were fixed with formaldehyde/glutaraldehyde to avoid agglutination, and were stored for sev-
Transplantation • Volume 84, Number 12S, December 27, 2007
Yung et al.
© 2007 Lippincott Williams & Wilkins
A FIGURE 1. Schematic overview of the ABO fluorescence-activated cell sorting (ABO-FACS) method employed. (A) Blood group A or B red blood cells (RBCs), and donor-specific and porcine RBCs were incubated with undiluted patient serum after fixation with formaldehyde/glutaraldehyde. After 30 min incubation at 4°C, unbound antibodies (Abs) were washed away and the binding of specific anti-A/B or porcine antibodies was detected by incubation with secondary isotype-specific fluorescein isothiocyanate–labeled Abs for another 30 min at 4°C. (B) Finally, anti– blood group or antiporcine Ab presence was detected by flow cytometry.
B
4oC, 30 min
S21
4oC, 30 min
washing
washing FACS sample acquisition
Anti-A/B or porcine Ab
Unbound Ab RBC from A/B blood type donor/ third-party or porcine
Secondary anti-human IgG or IgM FITC
TABLE 1. Anticarbohydrate antibodies absorbed by ABO Glycosorb A columns after immunoadsorption Mean fluorescence intensity ratio Type A RBC/ protein concentration
Donor RBC/ protein concentration
Type B RBC/ protein concentration
Swine RBC/total protein
IADS eluate
Protein in eluates (mg/mL)
IgM
IgG
IgM
IgG
IgM
IgG
IgM
IgG
2 6 9 11 14
0.23 0.25 0.14 0.14 0.08
324.7 254.9 247.4 118.7 611.2
642.5 461.1 209.0 207.5 671.4
306.7 263.9 338.8 115.6 385.2
1,268.5 997.8 473.4 764.3 1,118.7
291.9 268.9 208.3 143.1 259.4
801.9 298.0 173.8 279.9 327.6
950.2 682.4 707.0 306.1 1,290.9
494.8 226.6 159.8 141.9 283.4
After each IADS, the protein content of the Glycosorb A columns was eluted by collecting a total of 10 different fractions. The fraction with the highest protein concentration was used for anti-A/B and antiporcine Ab determinations by ABO fluorescence-activated cell sorting (ABO-FACS). The Ab levels, determined by ABO-FACS, were divided by the total protein concentration of the eluate. IADS, immunoadsorption; RBC, red blood cells.
eral months at ⫺80°C. As shown in Figure 1, undiluted serum was incubated with fixed RBCs, and binding of Ab was measured by indirect flow cytometry using secondary isotype-specific fluorescein isothiocyanate (FITC)-labeled Ab (mouse antihuman IgG; Zymed Laboratories, Inc., San Francisco, CA; goat antihuman IgM; Sigma Chemical Co., Basel, Switzerland). Human AB and O sera were used as negative and positive controls, respectively. To exclude binding of irregular RBC Abs, commercially available O RBCs expressing a panel of different non-ABO RBC Ags were pooled and included in all assays (screening test for irregular RBC Abs; Immucor, Inc., Norcross, GA). To compare the levels of anti-A/B Ab binding, the geometric mean fluorescence intensity ratios (MFIRs) were calculated by dividing the mean fluorescence intensity of the sera of interest by the mean fluorescence intensity of the negative control. Agglutination and Indirect Antiglobulin Testing Serially diluted serum of the recipient was incubated with a 1% suspension of donor RBCs (stabilized in ID CellStab; DiaMed AG, Cressier sur Morat, Switzerland) in 0.9% sodium chloride solution for 5 min at room temperature in glass tubes. After centrifugation, agglutination was judged positive if the RBCs remained clotted in the tube upon gentle
shaking and the IgM titers were expressed as the highest positive dilution (16). An indirect antiglobulin test was performed to determine IgG titers. The same tubes were incubated for 30 min in a 37°C water bath. After three washing steps, two drops of polyspecific rabbit anti-human globulin (DiaClon Coombs-serum green; DiaMed AG) were added to the tubes. The IgG titers were judged positive if the RBCs remained clotted, and were indicated as the highest positive dilution. All experiments were performed in duplicate. Column Elution After each IADS, performed according to the Karolinska protocol (5), the contents of the 15 Glycosorb A columns (Glycorex Transplantation AB, Lund, Sweden) were eluted with 100 mL of 0.1 M citric acid (pH 3) and neutralized with Tris-base buffer as a modification of Andrew and Titus’ protocol (17). The protein concentration was estimated with the BCA protein assay kit (Pierce Biotechnology Inc, Rockford, IL) using bovine serum albumin as a standard. Anti-A/B IgM and IgG Abs were measured by ABO-FACS in the eluted fraction with the highest protein concentration. Values in Table 1 show the ratio between the MFIR of Ab binding and the total protein concentration.
S22
Transplantation • Volume 84, Number 12S, December 27, 2007
months posttransplantation, the patient is doing well with normal graft function. Anti-A/B/porcine IgM and IgG were measured before and after each IADS session as well as from the eluates of the Glycosorb A columns by agglutination, IAT, and ABO-FACS. Initial antidonor agglutination titers were 1:64 for both IgM and IgG. As shown in Figure 2, anti-A IgM and IgG were successfully removed by each IADS treatment, but increased again overnight. After the first week of IADS, the anti-A-IgG titer remained ⬎1:8 and consequently, the transplantation was postponed. After 14 IADS sessions, the antidonor IgM agglutination titer was reduced from 1:64 to 1:2 and the IgG titer from 1:64 to 1:4. By contrast, the antidonor IgM MFIR, as determined by ABO-FACS, was no longer measurable after eleven IADS sessions, while the agglutination titer was still 1:16. The antidonor IgG MFIRs were below the detection
RESULTS AND DISCUSSION A 56-year-old male patient with blood group O Rh⫹, who suffered from diabetic renal failure, received a kidney from a blood group A1 Rh⫹ female living donor according to the Karolinska protocol (5). Briefly, rituximab was administered 1 month before the planned transplantation and 14 IADS sessions in total were performed using Glycosorb A columns in order to reduce the anti-A IgG titer to ⬍1:8. Seven days after transplantation, an additional IADS treatment was performed because of increasing anti-A IgG titers in the absence of clinical evidence for humoral rejection. Two months after transplantation, a protocol biopsy showed a Banff IIa rejection, which was successfully treated with high-dose corticosteroids. Neither Ig deposition nor a recipient-type endothelial chimerism was detected by immunohistochemistry in this biopsy (data not shown). Currently, 7
A
B
IgM
140 Rituximab
130
1024
256
MFIR
64
80
32
70
16
60
8
50
10
64
9 8
32
7 16
6 5
40 30 20
4
4
2
3
0.5
0 -59
-40-39
-17-16-15
-13
-11-10 -9 -8 -7 -6
-4 -3 -2
-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Third party anti-A IgM
8 4
2
1
10
2
1
Days
0
1 -59
Agglutination titer
-40-39
-17-16-15
-13
-11-10 -9 -8 -7 -6
-4 -3 -2
-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Third party anti-A IgG
Donor anti-A IgM
Indirect agglutination
Donor anti-A IgG
C
D
160
16
160 Rituximab
150
IVIG
140
130
120
120 Transplantation
90
90
80
80
70
14
IADS
13
12
12
11
11
Transplantation
10
10
8
70
7
7
60
60
6
6
50
50
5
5
40
40
4
4
30
30
3
3
20
20
2
2
10
10
1
0
0 -59
-40-39
-17-16-15
Anti-B IgM
-13
-11-10 -9 -8 -7 -6
-4 -3 -2
-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Anti-Porcine IgM
MFIR
9
8
MFIR
9
Days
MFIR
100
15
IVIG
13
110
100
Rituximab
14
140
IADS
16
15
150
130 110
MFIR
128
Transplantation
11
Agglutination titer
90
256
12
MFIR
128
Transplantation
512
IADS
13
Agglutination titer
100
IVIG
14
IADS
110
1024 Rituximab
15
512
120
IgG
16
IVIG
1
0
0 -59
-40-39
-17-16-15
Anti-B IgG
-13
-11-10 -9 -8 -7 -6
-4 -3 -2
-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Anti-Porcine IgG
FIGURE 2. Measurement of anti-A/B and antiporcine Ab in an A to O ABO-incompatible living donor kidney transplantation recipient. Anti-A IgM/IgG (A and B); anti-B and antiporcine IgM/IgG (C and D). Anti-A IgM agglutination titers (right y-axis in A), indirect antiglobulin test (IAT) IgG titers (right y-axis in B), and mean fluorescence intensity ratios (MFIRs), as measured by ABO fluorescence-activated cell sorting (ABO-FACS) (left y-axis in A and B), are depicted over time (x-axis), using donor red blood cells (RBCs) (stars) and third-party A RBCs (circles). Anti-B (squares) and antiporcine Ab (triangles) agglutination titers and MFIR values are shown in Figure 2C (IgM) and D (IgG). The dotted lines represent the negative threshold values of the ABO-FACS assay. Serum samples were collected twice daily, before and after the IADS. Arrows indicate the day of IADS sessions, rituximab, intravenous immunoglobulin administration, and the day of transplantation. IVIG, intravenous immunoglobulin.
Yung et al.
© 2007 Lippincott Williams & Wilkins
limit after five IADS sessions. This difference between the standard assays and the ABO-FACS suggests that ABOincompatible kidney transplantations could be performed after fewer IADS sessions. However, despite the previous demonstration of a close correlation between the MFIR obtained by ABO-FACS and standard agglutination titers, a formal comparison of the sensitivity of the two assays remains to be conducted. The observed differences between agglutination titers and ABO-FACS MFIR values, in particular the higher titers obtained for IgM by agglutination, may be due either to a lower sensitivity of the ABO-FACS or to unspecific agglutination mediated by polyreactive IgM. The Karolinska and the Freiburg im Breisgau protocols for ABO-incompatible kidney transplantations both use donor RBCs for the agglutination assay and IAT (18, 19). The rationale behind using donor RBCs, which is logistically more demanding, is based on the assumption that antidonor titers more accurately reflect the risk of rejection, but this has not been clearly proven. The present case shows similar results using third-party and donor A RBCs by ABO-FACS, suggesting that measurements using donor RBCs may not be necessary (Fig. 2). However, this comparison was only performed in a single patient, warranting additional assays for confirmation. In addition to donor-specific IgM/IgG, we also measured anti-B and antiporcine Ab by ABO-FACS. It is noteworthy that the anti-B and antiporcine IgM were also absorbed by the Glycosorb A columns (Fig. 2C), raising questions about their specificity (20). Similar to the pattern observed with anti-A IgM, anti-B and antiporcine IgM decreased after each IADS, though not to the same extent. By contrast, after an initial decrease in anti-B IgG, the Ab remained stable in the further pretransplant course and increased after transplantation. Antiporcine IgG remained stable throughout the procedure. To confirm that the decrease in anti-B and antiporcine Ab in the patient’s serum was due to column absorption, the levels of anti-A/B and antiporcine Ab were analyzed in the corresponding column eluates. In Table 1, five representative samples of a total of 15 eluates are displayed. The results demonstrate not only the effective removal of anti-A by third-party and donor RBCs, but also of anti-B and antiporcine IgM/IgG. The ratio between MFIR levels for IgM and IgG in the eluates differed considerably from the ratio in the serum, potentially due to a partial IgM pentamer degradation after acidic elution. In contrast to our findings, levels of Abs against another carbohydrate Ag (pneumococcus) and a protein Ag (tetanus) were not affected by IADS in another study (18). These conflicting results may be explained by the close structural relationship between ␣Gal and ABO Ag and the ensuing crossreactivity of anti-B and anti-␣Gal Ab to the synthetic A Ag attached to the Glycosorb A column (12). Accordingly, Rydberg et al. originally reported that O serum absorbed with Glycosorb A columns also retains anti-B Ab (21). In summary, the comparison of ABO-FACS results with agglutination and IAT titers showed that 1) antidonor IgM/IgG was detectable for longer time periods by agglutination/IAT, suggesting unspecific binding in these assays; 2) binding of anti-A Ab to donor and third-party A RBCs was similar; and 3) IADS did not selectively remove donor-spe-
S23
cific anti-A Ab, but also removed anti-B and antiporcine Ab, which was confirmed by the detection of anti-A/B and antiporcine Ab in column eluates. These initial results, as well as the analysis of anti-A/B Ab by ABO-FACS in additional cases of ABO-incompatible kidney transplantation, will allow a better understanding of the Ab kinetics and help to improve future IADS protocols. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8.
9. 10. 11.
12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
Segev DL, Simpkins CE, Warren DS, et al. ABO incompatible high-titer renal transplantation without splenectomy or anti-CD20 treatment. Am J Transplant 2005; 5: 2570. Sonnenday CJ, Warren DS, Cooper M, et al. Plasmapheresis, CMV hyperimmune globulin, and anti-CD20 allow ABO-incompatible renal transplantation without splenectomy. Am J Transplant 2004; 4: 1315. Takahashi K. ABO incompatible kidney transplantation. Philadelphia: Elsevier Science, 2001. Takahashi K, Saito K, Takahara S, et al. Excellent long-term outcome of ABO-incompatible living donor kidney transplantation in Japan. Am J Transplant 2004; 4: 1089. Tyden G, Kumlien G, Genberg H, et al. ABO incompatible kidney transplantations without splenectomy, using antigen-specific immunoadsorption and rituximab. Am J Transplant 2005; 5: 145. Winters JL, Gloor JM, Pineda AA, et al. Plasma exchange conditioning for ABO-incompatible renal transplantation. J Clin Apheresis 2004; 19: 79. Milland J, Sandrin MS. ABO blood group and related antigens, natural antibodies and transplantation. Tissue Antigens 2006; 68: 459. Lowe JB, Marth JD. Structures common to different types of glycans. In: Varki A, Cummings R, Esko J, Freeze H, Hart G, Marth JD, eds. Essentials of Glycobiology. New York: Cold Spring Harbor Laboratory Press, 1999: 211. Takahashi K, Saito K. Present status of ABO-incompatible kidney transplantation in Japan. Xenotransplantation 2006; 13: 118. Tanabe K, Takahashi K, Sonda K, et al. Clinicopathological analysis of rejection episodes in ABO-incompatible kidney transplantation. Transplant Proc 1996; 28: 1447. Norden G, Briggs D, Cockwell P, et al. ABO-incompatible live donor renal transplantation using blood group A/B carbohydrate antigen immunoadsorption and anti-CD20 antibody treatment. Xenotransplantation 2006; 13: 148. Galili U. Xenotransplantation and ABO incompatible transplantation: The similarities they share. Transfus Apher Sci 2006; 35: 45. Stussi G, West L, Cooper DK, Seebach JD. ABO-incompatible allotransplantation as a basis for clinical xenotransplantation. Xenotransplantation 2006; 13: 390. Stussi G, Huggel K, Lutz HU, et al. Isotype-specific detection of ABO blood group antibodies using a novel flow cytometric method. Br J Haematol 2005; 130: 954. Baumann BC, Stussi G, Huggel K, et al. Reactivity of human natural antibodies to endothelial cells from Galalpha(1,3)Gal-deficient pigs. Transplantation 2007; 83: 193. Brecher ME. Technical Manual. Bethesda, MD: American Association of Blood Banks, 2003. Andrew S, Titus J. Induction of immune responses. In: Coligan J, Kruisbeek A, Margulies D, Shevach EM, Strober W, eds. Current Protocols in Immunology. New York: Wiley, 1991: 2.7.4. Kumlien G, Ullstrom L, Losvall A, et al. Clinical experience with a new apheresis filter that specifically depletes ABO blood group antibodies. Transfusion 2006; 46: 1568. Donauer J, Wilpert J, Geyer M, et al. ABO-incompatible kidney transplantation using antigen-specific immunoadsorption and rituximab: A single center experience. Xenotransplantation 2006; 13: 108. Holgersson J, Gustafsson A, Breimer ME. Characteristics of proteincarbohydrate interactions as a basis for developing novel carbohydratebased antirejection therapies. Immunol Cell Biol 2005; 83: 694. Rydberg L, Bengtsson A, Samuelsson O, et al. In vitro assessment of a new ABO immunosorbent with synthetic carbohydrates attached to sepharose. Transpl Int 2005; 17: 666.
