Evaluation of inﬂ uence of IL-6 C-572G gene polymorphism and clinical factors on positive platelet antibody test

Background: Interleukin-6 ( ) promotes antibody production. The objective of this study was to investigate whether IL-6 C-572G single nucleotide polymorphisms (SNP) and clinical factors are associated with positive platelet antibody test. Materials and methods: Thirty platelet recipients with platelet antibodies (responders) and 20 platelet recipients without platelet antibodies (non-responders) were randomly selected. The -572 C>G (rs 1800796) SNPs in the promoter region of IL-6 gene were genotyped by polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) method. Solid phase red cell adherence assay (SPRCA) was used for platelet antibody detection. Results: Age, sex, percentage patients with benign diseases, and percentage of patients with homozygotes for the C allele at position -572 of the IL-6 gene were similar between responders and non-responders. Although the amounts of platelets pheresis transfused to patients with hematologic diseases were higher than those of non-hematologic diseases (47.2 ± 54.2 vs. 17.4 ± 13.8 units, p = 0.019), detection rate of platelet antibodies was lower in patients with hematologic diseases than that in patients with non-hematologic diseases (42.3% vs. 79.2%,


Introduction
Alloimmunization to human leukocyte antigens (HLAs) and to human platelet antigens (HPAs) may occur after exposure to nonself-antigens during transfusion or pregnancy. Platelet alloimmunization is one of several factors that can contribute to the platelet refractory state. Knowledge of predictors of alloimmunization may be useful for the prevention and management of platelet refractoriness.
Rhesus (Rh) antigens are not present on the platelet surface. However, platelet concentrates may contain enough production by promoting the B cell helper capabilities of CD4 + T cells through increased IL-21 production. IL-6 could therefore be a potential coadjuvant to enhance humoral immunity [5]. IL-6 inhibits TGF-β-induced regulatory T (Tregs) differentiation [6]. Tregs cells play a major role in restraining immune responses to maintain immune homeostasis [7]. A reduced Treg activity was observed in transfused mice that make alloantibodies (responders) [8].
IL-6 gene, located in 7p21, spans 5kb. IL-6 C-572G mutation in the promoter region is that cytosine (C) being substituted by guanine (G). The single nucleotide polymorphisms (SNP) with the presence of G at position -174 (rs1800795) within the 5' region of the IL-6 gene was found to be associated with increased donor-speci ic HLA antibody production after renal transplantation [9]. The -174C allele of the IL-6 promoter region was reported to be associated with systemic onset juvenile idiopathic arthritis [10], and susceptibility to rheumatoid arthritis in Europeans [11]. IL-6 -174C allele is rare in Chinese population [12,13]. Pan, et al. studied 232 individuals and found that 231 carried the wild type of -174G/ G; only 1 carried the GC genotype, and the CC genotype was not detected at all [12]. We previously studies 149 individuals and found that all individuals had GG at -174 position of IL-6 gene without any diversity [13]. But there are no published studies that have examined the association between IL-6 C-572G SNP and platelet antibody production.

Study subjects
The total number of healthy subjects and patients is 100: 50 healthy subjects, 30 with platelet antibody (responders) and 20 without platelet antibody (non-responders).
In 50 healthy subjects (21 women and 29 men; median age: 32.5 years, range: 20-58 years), they had no any active medical problem, or evidence of abnormal hepatic or renal function from screening laboratory test results in routine physical examination.
Cumulative amounts of platelets pheresis transfused in responders were calculated until the platelet antibody test. Blood components transfused to patients with hematologic diseases or malignancies were leukoreduced, but blood components for patients with non-hematologic diseases or non-malignant diseases were non-leukoreduced respecting the regulation of the reimbursement of leukoreduced components in Taiwan.
The study protocol was approved by the institutional review board of Taipei Veterans General Hospital. All subjects provided written informed consents.

Determination of platelet antibodies
The solid phase red cell adherence assay technique with the MASPAT kit (Sanquin, Amsterdam, The Netherlands) was used to detect antiplatelet antibodies (against HLA class I and HPAs) in patient serum. A positive or weak positive reaction indicates the presence of platelet and/or HLA speci ic antibodies in the serum.

Genomic DNA extraction and genotyping of IL-6
Genomic DNA was extracted from blood samples with Puregene DNA isolation kit (Gentra System, Minneapolis, MN, USA). The -572 C>G (rs 1800796) SNPs in the promoter region of IL-6 gene were genotyped by polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) method as previously described [13,14].

Statistics
Deviation from Hardy-Weinberg equilibrium was tested by using the chi-square test for goodness of it. For comparison of categorical variables between groups, the Pearson's chisquare test with Fisher's exact test was performed. The Mann-Whitney U test was used to compare numeric variables between groups. The differences between groups were considered signi icant if p -values of two-tails were less than 0.05. Statistical analysis was performed using IBM SPSS Statistics, Version 20 (IBM Corp. Armonk, NY, USA).

Patient characteristics
Age, sex, and percentage of patients with malignant diseases in responders were similar to those in nonresponders. Percentage of patients with hematologic diseases in responders was lower than that in non-responders (36.7% vs. 75%, p = 0.01, Table 1).

Association between hematologic diseases and platelet antibody production
No signi icant differences regarding baseline characteristics, such as age, sex percentage of patients with malignant disease, or percentage of patients with G positive genotype at position -572 of the IL-6 gene was noted between the hematologic and the non-hematologic group, Table 3. Total number of transfused platelets pheresis were higher in patients with hematologic diseases than those of nonhematologic diseases (47.2 ± 54.2 vs. 17.4 ± 13.8, P = 0.019) ( Table 3), but detection rate of platelet antibodies was lower in patients with hematologic disease than that in patients with non-hematologic diseases (42.3% vs. 79.2%, p = 0.01) ( Table 1). Compared to patients with hematologic diseases, patients with non-hematologic diseases had 1.871 (95% CI: 1.142 -3.065) fold relative risk of platelet antibody production.

Discussion
Our data found that percentage of patients with hematologic diseases in responders (with platelet antibody production) was lower than that in non-responders. Whereas percentage patients with malignancies in responders was similar to that in non-responders (40% vs. 60%, p = 0.248), Table 1.
The ratio of hematologic malignancies to non-hematologic malignancies in responders is lower in responders than that in non-responders (1:3 vs. 3:1, p = 0.039).

Frequencies of genotypes and alleles of -572C/G SNPs of IL-6 gene
The distribution of genotypes for IL-6 -572C>G (rs1800796) polymorphism did not differ from Hardy-Weinberg proportion in either patients or heathy volunteers. The frequencies of genotypes and alleles of IL-6 -572 C>G polymorphism was similar between patients and healthy subjects ( Table 1).
In the responder group, percentage of the recipients with a G-positive (CG + GG) genotype was the same with that of the non-responder group, (40% vs. 40%, p = 1.0) ( Table 1).
Age, sex, percentage of patients with benign disease, percentage of patients with hematologic disease, and amounts of platelet transfused did not differ between the CC genotype patient age, female sex, and IL-6 C-572G gene polymorphism are not risk factor for platelet antibody production in platelet recipients. In our study, age of healthy control subjects was younger than patients (responders and non-responders), but age of responders and non-responders was similar (p = 0.804). So age had no impact on the data analysis for platelet antibody production.
The prevalence of HLA class I antibodies was 2.27% (2/88) in donors without pregnancy or transfusion history [15]. Schnaidt, et al. tested sera of 500 female blood donors after pregnancy for platelet-speci ic antibodies (HPA-1, 3, 5) using the monoclonal antibody immobilization of platelet antigens (MAIPA) assay. Twenty-one sera (4.2%) were found to be positive: four anti-HPA-1a, one anti-HPA-5a and 16 anti-HPA-5b. Sera were positive up to 30 years after the last pregnancy [16]. Koerner and colleagues analyzed serum antibodies to platelet-speci ic glycosphingolipids (GSL) which antibodies were detected only in sera of immune thrombocytopenic purpura (ITP) patients. They found one half (6/12) of patients with ITP had antibodies bound to plateletspeci ic GSL. Sera from none of 10 patients with nonimmune thrombocytopenia, and only 1 of 18 normal subjects gave positive reactions with the platelet-speci ic GSL group [17].
Cytokines are important regulators of immune responses against foreign antigens. The response to foreign antigens is mainly regulated by CD4 + T-helper cells (Th), which can be subdivided into the distinct subsets Th1 and Th2 based on the cytokines they produce. The Th2 subset help B cells to proliferate and differentiate and is associated with humoraltype immune responses. IL-6 promotes Th2 differentiation   [18]. IL-6 plays a major role in in lammatory responses and antibody production [19]. IL-6 was found to have a function in restricting early Friend virus replication [20]. We have previously found that the IL-6 C-572G gene polymorphism is signi icantly associated with anti-E antibody (anti-E) production, with the allele G as a risk allele [13].
Friedman, et al. found a high rate of alloimmunization to both RBC (30%) and platelet antigens (85%) in transfused sickle cell disease (SCD) patients, but detected no signi icant association between RBC and platelet alloimmunization. This suggests that responders to RBC antigens are not more likely to respond to the antigenic determinants responsible for platelet alloimmunization [21]. Verduin, et al. showed that women with sickle cell disease were at a higher relative risk (27.0%) of developing RBC alloimmunization compared to men with the same disease. In patients without hemoglobinopathy, the risk of RBC alloimmunization appeared to be similar among men and women [22]. This is likely explained by more exposure to immunizing events through pregnancy and/or transfusions in females with sickle cell disease. In our study, female patients tend to have an increased relative risk on platelet alloantibodies compared with men (53.3% vs. 30%, p = 0.103). Sharpe, et al. found that the HLA-alloimmunization rate was different according to gender: 22% for males (4/18) vs. 67% for females (10/15) [23].
In this study, total platelet requirements in hematologic patients were higher than those in non-hematologic patients. This may be explained by that hematologic patients often develop thrombocytopenia as a consequence of the underlying disease itself and therapy, and platelet transfusion is required for the prevention or treatment of hemorrhagic manifestations in patients with thrombocytopenia. Wang, et al. found that anti-HLA and anti-HPA alloantibodies were found in 114 of 204 (55.88%) patients with platelet refractoriness, including 110 (96.49%) with anti-HLA alloantibodies only, 2 (1.75%) with anti-HPA alloantibodies only and 2 (1.75%) with both anti-HLA and anti-HPA alloantibodies (anti-HPA-3a and anti-HPA-5b) [24]. Kiefel, et al. studies serum samples from 252 patients with hematologic or oncologic diseases, and platelet antibodies were detected in the sera of 113 patients (44.8% of 252) [25]. In the present study, the detection rate of platelet antibodies in hematologic patients was 42.3%, which was lower than that of patients with non-hematologic diseases (79.2%, p = 0.01).
Platelets express HLA class I antigens, but platelets do not express HLA class II antigens [26]. Primary HLA alloimmunization is dependent on recognition of both HLA Class I and Class II antigenic differences. The presence of viable cells bearing Class II antigens such as lymphocytes and antigen-presenting cells is necessary to stimulate primary HLA alloimmunization. If ways were found either to remove the Class II-carrying leukocytes physically or to inactivate them, HLA alloimmunization could be prevented. This method will not, however, prevent the formation of platelet-speci ic alloantibody, which is due to transfusion of incompatible platelet antigens [27].
In a meta-analysis, Vamvakas combined the results of 8 randomized controlled trial and found that a statistically signi icant bene it from white cell reduction in preventing HLAalloimmunization [28]. Seftel, et al. reported that universal prestorage leukoreduction reduces platelet alloimmunization in multiply transfused patients, patients undergoing stem cell transplantation as well as acute leukemia patients receiving aggressive chemotherapy [29]. In our study blood components for patients with hematologic diseases were leukoreduced, leukoreduction might be responsible for a lower rate of alloimmunization in patients with hematologic disease as compared with patients with non-hematologic disease.
Friedman, et al. [21]. suggested that leukodepletion of blood products might be less ef icacious in immunocompetent SCD patients than in cancer and BMT patients who are immunocompromised for prevention of platelet alloimmunization. Intensive chemotherapy may suppress the production of antibody. Schonewille, et al. found that patients who underwent intensive chemotherapy formed antibodies at a much lower rate than other patients [30]. Reinhardt, et al. found that intensive chemotherapy in children with malignancies causes partial immune de iciency, including long-term impairment of humoral immunity [31]. Asfour, et al. found that transfusion of RhD-positive blood components to Rh-negative patients with hematologic cancers, who have received RhD-negative bone marrow and/or peripheral blood stem cells, are at low risk of developing RhD antibodies [32]. Schonewille, et al. studied alloimmunization after blood transfusion in patients with hematologic and oncologic diseases and found that patients who underwent intensive chemotherapy formed antibodies at a much lower rate than other patients [30]. Arora, et al. found that the rate of alloimmunization was higher in patients who had solid cancers (22.6%) or myelodysplastic syndrome (23%) compared with those who had other hematologic malignancies (7%) [33]. The intensity of chemotherapy for patients with hematologic malignancies was higher than that for patients with nonhematologic malignancies. In our study, the percentage of patients with malignancies were similar between responders and non-responders (40% vs. 60%, p = 0.248), but the ratio of hematologic malignancies to non-hematologic malignancies were lower in responders than non-responders (1:3 vs. 3:1, p = 0.039). Because blood components transfused to patients with hematologic diseases and malignant diseases were leukoreduced, we can assumed that higher intensity of chemotherapy might contribute more than leukoreduction on reducing platelet antibody formation.
In conclusion, our inding suggest that there is no association between IL-6 C-572G gene polymorphism and positive reactivity in solid phase platelet antibody detection method in platelet recipients. Higher intensity of chemotherapy might contribute more than leukoreduction on reducing platelet antibody formation.