Complement activating ABO anti-A IgM/IgG act synergistically to cause erythrophagocytosis: implications among minor ABO incompatible transfusions

EH-PT had an IgM titer of 256 and an IgG titer of 1024. All 30 donor plasma had IgM titers ≤ 64. Thirteen donor plasma had IgG anti-A titers ≤ 128 while 17 were ≥ 256. Ten out of 30 plasma samples activated complement C3b and were independent of the IgG titer (Table 1) [18]. IgG anti-A titers for the donor plasma causing the minor ABO transfusion reaction and sera H1–H5 are summarized in Table 2.

Functional studies were performed with monoclonal BRAD3 anti-D at a concentration of 1:800 and BRAD5 at a concentration of 1:100. BRAD3 sensitization caused significant erythrophagocytosis with all the 3 Rh phenotypes (> 65%) while BRAD5 sensitization caused lower phagocytosis (≤ 6%) (Additional file 1: Figure S1). IVIG inhibited monocyte phagocytosis, with a 50 percent inhibitory concentration (IC50) of 5.3 ± 0.4 µg/mL (Additional file 1: Figure S2). IgG1 (IC50 = 6 µg/mL) and IgG3 (IC50 = 2 µg/mL) inhibited phagocytosis more efficiently than IgG4 (IC50 = 79 µg/mL) and IgG2 (IC50 = 260 µg/mL) (Additional file 1: Figure S3). Effect of diluent on monocyte mediated erythrophagocytosis was measured. Percent erythrophagocytosis of CFDA-SE stained RBCs was observed to remain same (~ 90%) when 1/800 BRAD3 prepared in different diluents 0.2% FBS/PBS, Plasma, and Serum was used for the sensitization of RBCs (Additional file 1: Figure S4).

The following samples were evaluated in this study: (1) plasma from a single Group O donor that caused extravascular hemolysis in a Group A patient of a minor ABO incompatible apheresis platelet transfusion (EH-PT), (2) random Group O plasma (n = 30) from healthy whole blood donors, and (3) selected Group O sera (n = 10) from outpatients being evaluated for possible ABO incompatible kidney transplantation. The plasma from healthy donors had known IgM/IgG anti-A titers, some with the ability to activate complement as previously published [18]. The serum samples had known IgG anti-A titers and demonstrated anti-A hemolysis at low titer when tested with freshly clotted serum. Samples demonstrating hemolysis were selected since ‘hemolysin activity’ has been associated with ‘immune’ anti-A rather than naturally occurring anti-A [28].

IgM and IgG anti-A titers were evaluated using buffered and anti-IgG gel agglutination cards (Ortho Clinical Diagnostics, Pompano Beach, FL, USA), respectively according to the manufacturer’s instructions. To evaluate the effect of direct agglutination by pentameric IgM, the serum samples with high anti-A titers were treated with 1.1 M 2-mercaptoethanol (2-ME) in a 10:1 ratio at 37 °C for 15 min to assess hemagglutinin titers using buffered and anti-IgG gel agglutination cards. Complement activation of anti-A sera was evaluated at a dilution of 1:100 (vol:vol) according to the previously published protocol [18]. Briefly, a 5% Group A1 + RBC suspension was incubated at 37 °C for 30 min with 200 µL 1:100 serum and 200 µL pooled freshly clotted Group A serum from healthy donors as a source of complement. Sensitized RBCs were washed and checked for complement C3 activation using anti-C3b/d (Gamma-clone, Immucor, Norcross, GA, USA). IgG titers ≥ 256 were deemed ‘high titer’ for the purposes of this publication.

Group A1 + RBCs were used in the study. Rh-positive (R1r) RBCs were used in initial studies to establish the MSA. A 5% RBC suspension in 0.2% bovine serum albumin/phosphate buffered saline (pH 7.4) was labeled with 6 mM 5(6)-Carboxyfluorescein diacetate N-succinimidyl ester (CFDA-SE; Sigma Aldrich, St. Louis, MO, USA). The labeling reaction was stopped with heat inactivated fetal bovine serum (FBS; Thermo Fisher Scientific, Waltham, MA, USA). The reaction mix was washed and the RBC pellet was resuspended to 5% with 0.2% FBS/PBS (pH 7.4).

Based on initial observations from 30 Group O plasma [18], the samples were divided into 4 groups: (1) low IgG titer, non-complement activating, (2) high IgG titer, non-complement activating, (3) low IgG titer, complement activating, and (4) high IgG titer, complement activating. The patient sera with hemolysin activity were divided into 2 clusters (1) low IgG titer and (2) high IgG titer. Representative donor plasma and patient sera from each cluster (n = 12, highlighted in Table 1) were diluted 1:100 and used to sensitize Group A1 + red cells at 37 °C for 30 min with and without fresh Group A serum. All sensitized RBCs were washed and checked for IgG sensitization and complement C3 activation using anti-IgG and anti-C3b/d (Gamma-clone, Immucor, Norcross, GA, USA), respectively. An aliquot of sensitized, washed RBCs resuspended at 1% in monocyte media (CO₂ independent media containing 10% FBS and 1% l-Glutamine; Thermo Fisher Scientific, Waltham, MA, USA) was used to evaluate monocyte erythrophagocytosis.

Peripheral blood mononuclear cells (PBMCs) from a healthy donor were isolated from freshly collected ACD whole blood using density gradient medium (Stemcell Technologies, Vancouver, BC, Canada). Monocytes were purified from PBMCs by immunomagnetic CD14-positive selection (Stemcell Technologies, Vancouver, BC, Canada), frozen in 10% dimethyl sulfoxide/FBS (Thermo Fisher Scientific, Waltham, MA, USA) at − 80 °C and then thawed on the day of use. Before performing MSA, monocytes were thawed rapidly at 37 °C, washed once using saline, and resuspended in monocyte media. The cells were counted and assessed for viability using counting beads (Thermo Fisher Scientific, Waltham, MA, USA) and propidium iodide (Stemcell Technologies, Vancouver, BC, Canada). To measure phagocytosis, 50 µL monocytes (0.5–1 × 10⁵) were incubated with 60 µL 1% sensitized RBCs in 190 µL monocyte media (total reaction volume = 300 µL) in the dark at 37 °C for 30 min. Fluorescent labeled unsensitized RBCs served to evaluate as a negative control for erythrophagocytosis. Phagocytosis was stopped by one wash with saline and non-phagocytosed RBCs were lysed with 4 °C ammonium chloride for 5 min followed by one wash in saline. Flow cytometry was performed on FACSCalibur (BD Biosciences, CA). Data was analyzed using CellQuest Pro software (BD Biosciences, Franklin Lakes, NJ, USA). Increase in fluorescence in the FL1 channel (517 nm) was recorded as a positive signal for phagocytosis (Fig. 3). The phagocytic function of monocytes in suspension was evaluated by showing the effect of (1) Rh D antigen dosage on erythrophagocytosis using BRAD3 and BRAD5, (2) the dose–response inhibition by IVIG, and (3) the variable inhibitory effect of IgG subclasses (Additional file 1). Erythrophagocytosis ≥ 5% was deemed a significant observation based on reproducibility studies and the values obtained for the negative controls.

EH-PT and sera with high IgG titers (N = 5) were evaluated for IgG subclass after 2-ME treatment. 2-ME treatment was needed as high IgM in the samples was causing direct agglutination thereby obscuring the evaluation of IgG subclass at 1:50 dilution. A 1:50 diluted 2-ME treated sera were used to sensitize A1 + RBCs at 37 °C for 30 min. The reactions were washed and evaluated for the subclass using 1:50 diluted Anti-IgG1, -IgG2, -IgG3, and -IgG4 (Sigma Aldrich, St. Louis, MO, USA).