Alloantibodies to a paternally derived RBC KEL antigen lead to hemolytic disease of the fetus/newborn in a murine model

Abstract

Exposure to nonself red blood cell (RBC) antigens, either from transfusion or pregnancy, may result in alloimmunization and incompatible RBC clearance. First described as a pregnancy complication 80 years ago, hemolytic disease of the fetus and newborn (HDFN) is caused by alloimmunization to paternally derived RBC antigens. Despite the morbidity/mortality of HDFN, women at risk for RBC alloimmunization have few therapeutic options. Given that alloantibodies to antigens in the KEL family are among the most clinically significant, we developed a murine model with RBC-specific expression of the human KEL antigen to evaluate the impact of maternal/fetal KEL incompatibility. After exposure to fetal KEL RBCs during successive pregnancies with KEL-positive males, 21 of 21 wild-type female mice developed anti-KEL alloantibodies; intrauterine fetal anemia and/or demise occurred in a subset of KEL-positive pups born to wild type, but not agammaglobulinemic mothers. Similar to previous observations in humans, pregnancy-associated alloantibodies were detrimental in a transfusion setting, and transfusion-associated alloantibodies were detrimental in a pregnancy setting. This is the first pregnancy-associated HDFN model described to date, which will serve as a platform to develop targeted therapies to prevent and/or mitigate the dangers of RBC alloantibodies to fetuses and newborns.

Figure 1

Smaller successive litters with fewer KEL-positive pups are born to wild-type females mated with KEL males. (A) Number of pups alive at birth, (B) number of pups alive at weaning, and (C) percentage of KEL-positive pups in first and third litters of wild-type females mated with KEL males. (D) Number of pups alive at birth, (E) number of pups alive at weaning, and (F) percentage of KEL-positive pups in first and third litters of control KEL females mated with wild-type males. (A-C) Data are a compilation of 40 total pregnancies; (D-F) data are a compilation of 16 total pregnancies. *P < .05.

Figure 2

A subset of pups born to multiparous wild-type females mated with KEL males are stillborn or pale. (A) Photograph of a representative pink KEL-negative and pale KEL-positive pup, hours after birth to a multiparous (third litter) wild-type female mated with a KEL male. (B) Representative KEL-specific PCR. (C) Flow cytometric analysis of the blood of the pups shown, with 1000 Trucount beads and RBCs gated. (D) TER119 positivity of gated RBCs. (E) Thiazole orange reticulocyte staining and (F) TER119 staining of the blood of the same pups. (G) Blood smear of representative KEL-negative and KEL-positive pups. (H) MRI of first (left) or third (right) pregnancies of representative wild-type females bred with KEL males.

Figure 3

KEL is expressed on all RBC precursors in the fetal liver and bone marrow of transgenic animals, in an RBC-specific fashion. (A-C) Fetal liver cells from KEL transgenic fetuses, with anti-KEL staining (solid) or secondary only staining (shaded) of RBC precursor populations (I = primitive erythroid progenitor cells, II = proerythroblasts and early basophilic erythroblasts, III = later basophilic erythroblasts, IV = chromatophilic and orthochromatophilic erythroblasts, and V = late orthrochromatophlic erythroblasts and reticulocytes). (D) Bone marrow from KEL transgenic animals, with anti-KEL staining (solid) or secondary only staining (shaded) of RBC populations (I = proerythroblasts, II = basophilic erythroblasts, III = polychromatic erythroblasts, IV = orthochromatic erythroblasts, V = reticulocytes, VI = mature RBCs). (E) Anti-KEL staining of peripheral blood cells. (F) Bone marrow and organs of a KEL-positive animal were evaluated for KEL expression by PCR. Results are from 2 independent experiments, with 1 to 2 mice per group.

Figure 4

Fetal RBCs can be detected in the maternal circulation after delivery, with anti-KEL RBC antibodies detectable in the serum of wild-type females after multiple pregnancies with KEL-positive males. (A) Unstained RBCs from a nulliparous wild-type female. (B) Unstained RBCs from a homozygous uGFP male. (C) Unstained RBCs from a representative wild-type female bred with a homozygous uGFP male, p.c. day 7, p.c. day 14, and hours after delivery. (D-E) Serum from nulliparous or multiparous animals was cross matched with TER119-positive RBCs. (F) Representative flow cross match against wild-type (shaded) or KEL (solid) RBCs, from a nulliparous and a multiparous female, with anti-mouse IgG as a secondary reagent. (G) Anti-KEL IgG in the serum of nulliparous or multiparous (3 pregnancies) females. (H) Anti-KEL IgG in the serum of females after 1, 2, or 3 deliveries; 42 total pregnancies are shown. *P < .05.

Figure 5

Maternal anti-KEL IgG crosses the placenta and binds to KEL RBCs of pups. (A) RBCs from KEL-positive and KEL-negative pups born to mothers immunized through pregnancy were evaluated for bound anti-KEL IgG by flow cytometric cross matching. (B) Serum from these same animals was evaluated for anti-KEL IgG in a direct antiglobulin test. Results are representative of 3 independent breeding experiments (n = 20 pups). *P < .05. (C) Anti-KEL subtypes in the serum of a representative alloimmunized mother compared with the serum of her KEL-negative pups. (D) Newborn liver cells from a representative KEL-negative and pale KEL-positive pup born to an alloimmunized mother, quantitated by Trucount beads (left column) or stained with anti-mouse Igs (right column). (E) Total number of pups and percentage of KEL-positive pups was evaluated in first and third litters of MuMT females bred with KEL males (n = 8 total pregnancies). P is not significant. No anti-KEL antibodies were detected in any MuMT female; error bars indicate standard deviation.

Figure 6

Transfused KEL RBCs are selectively cleared in alloimmunized animals, with recipient proinflammatory cytokine response. (A) KEL and wild-type RBCs were labeled with DiI and DiO, respectively, prior to transfusion into females alloimmunized through pregnancy or into control recipients. (B) Post-transfusion survival and recovery of KEL RBCs was determined by comparing a ratio of DiI KEL to DiO wild-type RBCs. (C) RBC bound IgG was evaluated by flow cytometry 10 minutes post-transfusion. (D-F) Serum cytokine responses in alloimmunized animals 90 to 120 minutes after a KEL RBC transfusion. Results are representative (A-C) or a compilation (D-F) of 3 experiments (n = 18 animals). *P < .05.

Figure 7

KEL RBC transfusions induce boostable anti-KEL alloantibodies, which result in adverse pregnancy outcomes. (A) Control animals were transfused 3 times with KEL RBCs, with anti-KEL IgG subtypes and IgM measured in pooled serum and compared with pooled serum from 12 females alloimmunized through pregnancy. (B) Anti-KEL IgG measured in transfusion recipients 2 weeks after each transfusion. (C) Naïve or transfused females were bred to KEL males, with total pups born enumerated. (D) percentage of pups alive at weaning evaluated, and (E) percentage of KEL-positive pups determined. (B) Results are representative of 3 independent experiments. (C-E) Results are a compilation of 37 pregnancies. *P < .05.