Life-threatening hemolytic anemia due to an autoanti-Pr cold agglutinin: evidence that glycophorin A antibodies may induce lipid bilayer exposure and cation permeability independent of agglutination

BACKGROUND: The hemoglobin of a 29-year-old man fell below 35 g/L over 5 days, despite 14 units of red blood cells (RBCs), due to an anti-Pr cold agglutinin (CA). His hemolytic anemia necessitated respiratory support in intensive care for 4 weeks. STUDY DESIGN AND METHODS: The hemolysis was investigated by the effects on blood group-compatible RBCs of this anti-Pr and an anti-I CA and of a rabbit anti-human glycophorin A (GPA) immunoglobulin G (IgG) antibody on Ca(2+) permeability and of phosphatidylethanolamine (PE) exposure. 1) The anti-Pr CA (in a plasmapheresis product from the patient) was absorbed and eluted from RBC ghosts and its immunophenotype was determined by agarose electrophoresis and immunofixation. 2) Ca(2+) permeability was measured by the response of Fluo-3-labeled RBCs to addition of external Ca(2+). 3) Exposed PE was measured with streptavidin-labeled biotinylated peptide Ro 09-0198 (cinnamycin). RESULTS: 1) The patient's anti-Pr CA was a polyclonal IgG. 2) The anti-Pr and the rabbit anti-human glycophorin IgG, but not an anti-I CA, rapidly increased Ca(2+)-dependent fluorescence upon addition of external Ca(2+) in a fraction (15%-25%) of RBCs that also became positive for cinnamycin. 3) Trypsin treatment of RBCs reduced the Ca(2+) influx due to the anti-Pr IgG, but neither trypsin nor neuraminidase changed the responses to the rabbit anti-human GPA IgG. CONCLUSIONS: The anti-Pr CA and rabbit anti-human GPA increased exposure of PE and increased membrane Ca(2+) permeability that may have caused hemolysis. The difference in the responses to these antibodies to enzyme treatment of RBCs suggests that they react with different epitopes on GPA.