Red blood cell (RBC) transfusion exposes recipients to hundreds of unmatched minor RBC antigens. This exposure can lead to production of alloantibodies that promote clinically significant hemolytic events. Multiple studies have reported an increased frequency of RBC alloimmunization in patients with autoimmunity. However, cellular and molecular mechanisms that underlie autoimmunity-induced alloimmunization have not been reported. Patients with systemic lupus erythematosus (SLE) have a high frequency of alloimmunization and express a type 1 interferon (IFNα/β) gene signature. Thus, we utilized the pristane-induced lupus mouse model to test the hypothesis that inflammation in lupus promotes RBC alloimmunization, and to examine the potential role of IFNα/β. Intraperitoneal injection of pristane, a hydrocarbon oil, led to autoantibody production, glomerulonephritis, and pulmonary hemorrhage in wild type (WT) mice. Pristane treatment significantly induced serum IFNα and expression of multiple interferon-stimulated genes (ISGs) in peripheral blood and peritoneal fluid cells, including inflammatory macrophages. Following transfusion with allogeneic RBCs expressing the KEL glycoprotein, pristane-treated WT mice produced significantly elevated levels of anti-KEL IgM and anti-KEL IgG, compared to untreated mice. Pristane induced comparable levels of inflammatory cells and cytokines in mice lacking the IFNα/β receptor (IFNAR1 −/−) or the IFNα/β-inducing transcriptions factors (IRF3/7 −/−), compared to WT mice. However, pristane-treated IFNAR1 −/− and IRF3/7 −/− mice failed to produce ISGs and produced significantly lower levels of transfusion-induced anti-KEL IgG, compared to WT mice. Thus, pristane induction of a lupus-like phenotype promoted alloimmunization to the KEL RBC antigen in an IFNα/β-dependent manner. To our knowledge, this is the first examination of molecular mechanisms contributing to RBC alloimmunization in a model of autoimmunity. These results warrant further investigation of the role of IFNα/β in alloimmunization to other RBC antigens and the contribution of the IFNα/β gene signature to the elevated frequency of alloimmunization in patients with SLE.

Background RBC transfusion can lead to the production of alloantibodies against RBC antigens and is a clinically significant issue in transfusion medicine. Patients with sickle cell disease have an increased risk for alloimmunization; 30-50% of SS patients have alloantibodies compared to 3-10% of other hospitalized patients. These alloantibodies can cause dangerous hemolytic transfusion reactions and limit the availability of compatible antigen-negative RBC products. This is of particular importance in SS patients, who commonly make alloantibodies against multiple RBC antigens and need regular transfusions to treat their disease. However, mechanisms underlying the increased frequency of alloimmunization in sickle cell patients are poorly understood. In previous studies, inflammation in the recipient has been shown to promote alloimmunization in both mice and humans. In mouse transfusion models, type 1 interferons (IFNα/β) and Interferon Stimulated Genes (ISGs) have been shown to promote alloimmunization. Other studies have shown that patients with inflammatory autoimmune diseases express an IFNα/β signature, which may contribute to the increased frequency of alloimmunization in these populations. Given the chronic inflammatory state in patients with sickle cell disease, we hypothesize that: SS patients may also have an IFN gene signature that may contribute to the increased frequency of alloimmunization. Methods To test this hypothesis, we initially measured the expression of the ISG, myxovirus resistance protein 1 (MxA), in the blood of previously-transfused patients (n=50) with SS disease (SS, n=13) and without SS disease (ββ, n=37) by whole blood immunoassay (ELISA). We then measured expression of another ISG, Siglec-1 (SS n=5, ββ=24), expressed on monocytes by flow cytometric analysis. To determine the degree to which ISG expression correlated with alloimmunization frequency, expression of MxA in non-alloimmunized patients was compared to the expression in patients with 1 or more alloantibodies. Statistical analysis of 2 groups was completed with a Mann-Whitney U test. Significance between 3 groups was determined using a Kruskal-Wallis test with a Dunn's post-test. Results SS patients had significantly elevated levels of MxA (mean ± standard error of the mean, SS MxA = 8.98 ng/mL ± 2.46) compared to control patients without SS (MxA = 1.25 ± 0.54, p<0.0001). (Figure 1 A). SS patients also had significantly elevated levels of Siglec-1 on blood monocytes, measured by flow cytometric mean fluorescence intensity (MFI, SS MFI = 132.72 ± 42.9, ββ MFI = 64.9 ± 6.17, p< 0.05). (Figure 1 B,C). For all 50 patients, including SS and ββ control patients, there was a trend toward elevated MxA expression in alloimmunized patients. Patients with 2 or more alloantibodies had significantly elevated MxA (MxA 8.16 ± 2.61), compared to non-alloimmunized transfused patients (MxA = 2.05 ± 1.65, p < 0.01) or patients with only 1 alloantibody (MxA = 1.18 ±0.48, p<0.01). There was no significant difference in MxA levels between patients with 0 and 1 alloantibody. Of the 13 patients with SS disease, only 2 patients lacked alloantibodies. (SS with 1 alloantibody, n=3, SS with 2 or more alloantibodies, n=8). Therefore, a correlation between MxA levels and alloimmunization in SS patients could not be assessed. Discussion Factors that contribute to RBC alloimmunization in sickle cell disease are poorly understood. In this study, we found that sickle cell patients had an increase in the expression of ISGs compared to other transfused patients. We also found that MxA levels are increased in patients that have 2 or more alloantibodies compared to patients without alloantibodies. These findings suggest the presence of an IFNα/β gene signature in patients with sickle cell disease. Further studies are needed to determine the relationship between interferon-stimulated responses in sickle cell patients and the increased frequency of alloantibody production. Disclosures No relevant conflicts of interest to declare.

Background: Red blood cell (RBC) alloimmunization can follow exposure to foreign RBC antigens during transfusion therapy and during pregnancy/delivery. Alloantibodies against RBC antigens can cause potentially fatal hemolytic transfusion reactions and hemolytic disease of the newborn. Recent studies indicate that inflammatory states, including autoimmunity, promote RBC alloimmunization. For example, patients with systemic lupus erythematosus (SLE) have an elevated risk of RBC alloimmunization. Approximately 50% of SLE patients have a pro-inflammatory type 1 interferon (IFNα/β) gene signature, defined as the expression of interferon stimulated genes (ISGs), that is associated with disease severity. Recent studies in mouse transfusion models indicate that IFNα/β significantly promotes RBC alloimmunization. Thus, we hypothesized that IFNα/β produced in a lupus mouse model would promote RBC alloimmunization following transfusion. Methods: To test this hypothesis, we utilized the pristane-induced lupus mouse model, known to manifest lupus autoantibodies and a lupus-like phenotype. Mice lacking the IFNα/β receptor (IFNAR1-/-), mice lacking transcription factors required for IFNα/β production (IRF3/7-/-) and wildtype (WT) mice were administered an intraperitoneal injection of pristane 14 days prior to transfusion with transgenic RBCs expressing the KEL antigen. At the time of transfusion, serum IFNα levels were measured by ELISA, and expression of interferon stimulated gene 15 (ISG15) was measured by quantitative PCR. The mean fluorescence intensity (MFI) of the anti-KEL IgM and IgG responses were measured by flow cytometry 5 days and 7-28 days, respectively, following transfusion. A Mann-Whitney U test and a Kruskall-Wallis test with a Dunn's post-test was used to compare 2 or more than 2 groups, respectively. Results: Pristane induced serum IFNα in WT mice (Day 14 post-pristane, mean IFNα=127.3 Units/ml ± 28.21 standard error of the mean, SEM, p<0.01), which was undetectable in untreated mice. Pristane treatment also resulted in a non-significant trend toward upregulated expression of ISG15, post-pristane ISG15 =2.94 ± 1.16 SEM; no pristane ISG15 = 1.4 ± 0.2 SEM). Following transfusion, untreated mice did not form anti-KEL IgG. However, pristane-treated WT mice produced significantly elevated levels of anti-KEL IgM and anti-KEL IgG in 3 out of 3 experiments (untreated IgM MFI = 79.1 ± 13.5, pristane-treated IgM MFI = 142.7 ± 21.6, p<0.01; untreated IgG MFI = 10.7± 4.6, pristane-treated IgG MFI= 203.9 ± 99.6, p<0.05, Figure 1). Pristane-treated IFNAR1-/- and IRF3/7-/- mice produced significantly lower levels of anti-KEL IgG (IFNAR1-/- MFI = 39.0 ± 35.6, p<0.05; IRF3/7-/- MFI = 36.1 ± 16.3, p<0.02) compared to WT mice (MFI = 361 ± 148). Conclusion: Pristane induction of a lupus-like phenotype promoted alloimmunization to the KEL RBC antigen in an IFNα/β-dependent manner. These results warrant further investigation of the role of IFNα/β in alloimmunization to other RBC antigens, and the investigation of the contribution of IFNα/β responses to increased frequency of alloimmunization in patients with SLE. Disclosures No relevant conflicts of interest to declare.

Red blood cell (RBC) transfusion exposes recipients to hundreds of unmatched minor RBC antigens. This exposure can lead to production of alloantibodies that can cause potentially fatal hemolytic events. Prior studies have reported that inflammatory states, including autoimmune disease, in the transfusion recipient can promote alloimmunization. Recent studies reported that pro-inflammatory type 1 interferons (IFNα/β) regulated RBC alloimmunization in multiple mouse transfusion models. Interestingly, patients with systemic lupus erythematosus (SLE) have an increased frequency of alloimmunization and express an IFNα/β gene signature. Thus, we tested the hypothesis that IFNα/β activation in lupus promotes RBC alloimmunization by utilizing the pristane-induced lupus mouse model. Pristane significantly induced serum IFNα and expression of multiple interferon-stimulated genes (ISG) in peritoneal fluid cells, including inflammatory macrophages. Following transfusion with allogeneic RBCs expressing the KEL antigen, pristane-treated WT mice produced significantly elevated levels of anti-KEL IgM and anti-KEL IgG, compared to untreated mice. Pristane-treated mice lacking the IFNα/β receptor (IFNAR1−/−) or IFNα/β-inducing transcription factors (IRF3/7−/−) mice failed to produce ISGs and produced significantly lower levels of anti-KEL IgG compared to WT mice. In conclusion, pristane induction of a lupus-like phenotype promoted alloimmunization to the KEL RBC antigen in an IFNα/β-dependent manner. These results warrant further investigation of the role of IFNα/β in alloimmunization to other RBC antigens and the contribution of IFNα/β responses to the increased frequency of alloimmunization in patients with SLE.

RBC transfusion can lead to the production of alloantibodies against RBC antigens. RBC alloimmunization can result in hemolytic events and severely limit the availability of compatible RBCs. Patients with sickle cell disease (SCD) have the highest rate of alloimmunization, compared to any other disease population. However, mechanisms underlying this increased frequency are poorly understood. Inflammation in the recipient has been shown to promote alloimmunization in mice and humans. In mouse transfusion models, type 1 interferons (IFNα/β) and interferon-stimulated genes (ISGs) have been shown to promote alloimmunization. Given the chronic inflammatory state in patients with SCD, we hypothesized that SCD patients may also have an IFNα/β gene signature that may contribute to the increased frequency of alloimmunization. To test this hypothesis, we measured the expression of multiple ISGs, including myxovirus resistance protein 1 (MxA) and Siglec-1 by whole blood immunoassay and flow cytometry, respectively. MxA and Siglec-1 levels were significantly elevated in SCD patients, compared to transfused controls. We then determined the degree to which ISG expression correlated with alloimmunization frequency by linear regression. There was a trend toward elevated MxA expression in patients with 1 or more alloantibodies. Patients with 2 or more alloantibodies had significantly elevated MxA compared to non-alloimmunized transfused patients. These findings suggest the presence of an IFNα/β gene signature in patients with SCD and warrant further investigation of the contribution of IFNα/β pathways toward the profoundly elevated frequency of RBC alloimmunization in patients with SCD.