Two new JK silencing alleles identified by single molecule sequencing with 20-Kb long-reads.

BACKGROUND: The Kidd blood group gene SLC14A1 (JK) accounts for approximately 20 Kb from initiation codon to stop codon in the genome. In genomic DNA analysis using Sanger sequencing or short-read-based next generation sequencing, it is difficult to determine the cis or trans positions of single nucleotide variations (SNVs), which are occasionally more than 1 Kb away from each other. We aimed to determine the complete nucleotide sequence of a 20-Kb genomic DNA amplicon to characterize the JK allelic variants associated with Kidd antigen silencing in a blood donor. STUDY DESIGN AND METHODS: The Jk(a-b-) phenotype was identified in this donor by standard serological typing. A DNA sample obtained from whole blood was amplified by long-range PCR to obtain a 20-Kb fragment of the SLC14A1 gene, including the initiation and stop codons. The fragment was then analyzed by Sanger sequencing and single-molecule sequencing. Transfection and expression studies were performed in CHO cells using the expression vector construct of JK alleles. RESULTS: Sanger sequencing and single-molecule sequencing revealed that the donor was heterozygous with JK*01 having c.276G>A (rs763262711, p.Trp92Ter) and JK*02 having c.499A>G (rs2298719, p.Met167Val), c.588A>G (rs2298718, p.Pro196Pro), and c.743C>A (p.Ala248Asp). The two JK alleles identified have not been previously described. Transfection and expression studies indicated that the CHO cells transfected with JK*02 having c.743C>A did not express the Jkb and Jk3 antigens. CONCLUSIONS: We identified new JK silencing alleles and their critical SNVs by single-molecule sequencing and the findings were confirmed by transfection and expression studies.

Genetic background of anti-Xga producers in Japanese blood donors.

BACKGROUND AND OBJECTIVES: The Xg blood group is composed of two antigens, Xga (XG1) and CD99 (XG2 and MIC2). The XG and CD99 are homologous genes located on pseudoautosomal region 1 of the X and Y chromosomes. The expressions of Xga and CD99 are co-regulated by a single nucleotide polymorphism (rs311103) in the GATA-1 binding region. Another mechanism of the Xg(a-) phenotype is the genomic deletion of approximately 114 kb, including the XG gene. Anti-Xga seems to be naturally occurring by detection in males who have never been transfused. MATERIALS AND METHODS: In this study, we identified 23 anti-Xga producers among 580,115 donors (0.004%). Additional 12 anti-Xga producers were also identified from a separate cohort. RESULTS: All 35 anti-Xga producers were male. Genomic DNA was obtained from 34 of 35 producers, and all 34 producers were confirmed to carry the XG-gene-deficient allele (XGdel). The breakpoints of all 34 producers were identical. The XGdel was also identified in 12 non-producers of anti-Xga among 860 donors who have no antibodies against RBCs, and the breakpoints were also identical with the anti-Xga producers. CONCLUSION: Our results will serve as the basis for a more complete understanding of Xg blood group polymorphisms.

Reduction of blood group A antigen on erythrocytes in a patient with myelodysplastic syndrome harboring somatic mutations in RUNX1 and GATA2.

BACKGROUND: Reduction of blood group ABO antigens on red blood cells (RBCs) is well known in patients with leukemias, and this reduction of ABO expression is strongly associated with DNA methylation of the ABO promoter. Previously, we reported a two-nucleotide deletion in RUNX1 encoding an abnormally elongated protein lacking the trans-activation domain in a patient with myelodysplastic syndrome (MDS) showing A-antigen loss on RBCs. This prompted us to investigate the underlying mechanism responsible for A-antigen reduction on RBCs in another patient with MDS. STUDY DESIGN AND METHODS: Screening of somatic mutations was carried out using a targeted sequencing panel with genomic DNA from peripheral blood mononuclear cells from the patient and eleven MDS controls without A- or B-antigen loss. DNA methylation of the ABO promoter was examined by bisulfite genomic sequencing. Transient transfection assays were performed for functional evaluation of mutations. RESULTS: Screening of somatic mutations showed missense mutations in RUNX1 and GATA2 in the patient, while no mutation was found in exons of those genes in the controls. There was no significant difference in ABO promoter methylation between the patient and the controls. Transient transfection experiments into COS-7 and K562 cells suggested that the amino acid substitutions encoded by those mutations reduced or lost the trans-activation potential of the ABO expression. CONCLUSION: Considering the discrepancy between the variant frequencies of these mutations and the ratios of the RBCs with A-antigens loss, the antigen reduction might be associated with these somatic mutations and hypermethylation of the ABO promoter.

A new antigen SUMI carried on glycophorin A encoded by the GYPA*M with c.91A>C (p.Thr31Pro) belongs to the MNS blood group system.

BACKGROUND: MNS is one of the highly polymorphic blood groups comprising many antigens generated by genomic recombination among the GYPA, GYPB, and GYPE genes as well as by single-nucleotide changes. We report a patient with red blood cell (RBC) antibody against an unknown low-frequency antigen, tentatively named SUMI, and investigated its carrier molecule and causal gene. STUDY DESIGN AND METHODS: Standard serologic tests, including enzyme tests, were performed. Monoclonal anti-SUMI-producing cells (HIRO-305) were established by transformation and hybridization methods using lymphocytes from a donor having anti-SUMI. SUMI+ RBCs were examined by immunocomplex capture fluorescence analysis (ICFA) using HIRO-305 and murine monoclonal antibodies against RBC membrane proteins carrying blood group antigens. Genomic DNA was extracted from whole blood, and the GYPA gene was analyzed by polymerase chain reactions and Sanger sequencing. RESULTS: Serologic screening revealed that 23 of the 541,522 individuals (0.0042%) were SUMI+, whereas 1351 of the 10,392 individuals (13.0%) had alloanti-SUMI. SUMI antigen was sensitive to ficin, trypsin, pronase, and neuraminidase, but resistant to α-chymotrypsin and sulfydryl-reducing agents. ICFA revealed that the SUMI antigen was carried on glycophorin A (GPA). According to Sanger sequencing and cloning, the SUMI+ individuals had a GYPA*M allele with c.91A>C (p.Thr31Pro), which may abolish the O-glycan attachment site. CONCLUSIONS: The new low-frequency antigen SUMI is carried on GPA encoded by the GYPA*M allele with c.91A>C (p.Thr31Pro). Neuraminidase sensitivity suggests that glycophorin around Pro31 are involved in the SUMI determinant.

Novel hybrid genes and a splice site mutation encoding the Sta antigen among Japanese blood donors.

BACKGROUND: The low-incidence antigen Sta of the MNS system is usually associated with the GP(B-A) hybrid molecule, which carries the 'N' antigen at the N terminus. The GP(A-A) molecule with trypsin-resistant M antigen has been found in a few St(a+) individuals. MATERIALS AND METHODS: Among Japanese blood donors, we screened 24 292 individuals for the presence of St(a+) with trypsin-resistant 'N' antigen and 193 009 individuals for the presence of St(a+) with trypsin-resistant M antigen. The breakpoints responsible for the Sta antigen were analysed by sequencing the genomic DNAs. RESULTS: A total of 1001 (4·1%) individuals were identified as St(a+) with trypsin-resistant 'N' antigen. Out of 1001 individuals, 115 were selected randomly for sequencing. Two novel GYP*Sch (GYP*401) variants with new intron 3 breakpoints of GYPA were detected in three cases. Twenty-five (0·013%) individuals were identified as St(a+) with trypsin-resistant M antigen. Five individuals had the GYP(A-ψB-A) hybrid allele; two of these five individuals were GYP*Zan (GYP*101.01), and the remaining three had a novel GYP(A-ψB-A) allele with the first breakpoint in GYPA exon A3 between c.178 and c.203. Nine individuals had a novel GYP(A-E-A) allele with GYPE exon E2 and pseudoexon E3 instead of GYPA exon A2 and A3. The 11 remaining individuals had a novel GYP(A-A) allele with a 9-bp deletion that included the donor splice site of intron 3 of GYPA. CONCLUSION: Our finding on diversity of glycophorin genes responsible for Sta antigen provides evidence for further complexity in the MNS system.

An unusual variant glycophorin expressing protease-resistant M antigen encoded by the GYPB-E(2-4)-B hybrid gene.

BACKGROUND AND OBJECTIVES: MNS is a highly polymorphic blood group comprising 49 antigens recognized by International Society of Blood Transfusion, some of which may have been generated by genomic recombination among the closely linked genes GYPA, GYPB and GYPE. The GYPE gene has an almost identical sequence to GYPA*01 allele in exon 2 (99% homology), which accounts for M antigen. We investigated an unusual glycophorin molecule with protease-resistant M antigen. METHODS: Blood samples were screened by an automated blood typing system (PK7300) using bromelain-treated red blood cells (RBCs) and murine monoclonal anti-M. The M-positive RBC samples were analysed by immunoblotting using anti-M as the primary antibody. GYPA, GYPB and GYPE genes were analysed by polymerase chain reaction (PCR), cloning and sequencing using reticulocyte mRNA and genomic DNA. RESULTS: Serological tests and immunoblotting revealed that 103 of the 193 009 individuals (0·0534%) expressed protease-resistant M-active glycophorin having a molecular weight of 20 kDa. All the 103 individuals were S+ s- or S- s+. When reticulocyte mRNA from the individuals with M-active glycophorin (20 kDa) was examined by PCR and cloning followed by sequencing, a novel GYPE-B hybrid transcript was identified. Long-range PCR and sequencing using genomic DNA revealed that the individuals had a GYPB-E(2-4)-B hybrid gene. This hybrid gene was predicted to encode a 59-amino-acid mature glycoprotein that expresses no S or s antigens CONCLUSIONS: The prevalence of the M-active glycophorin (20 kDa) in the Japanese population is 0·0534%. This glycophorin is predicted to be a 59 amino acids polypeptide encoded by the novel GYPB-E(2-4)-B hybrid gene.

RUNX1 mutation in a patient with myelodysplastic syndrome and decreased erythrocyte expression of blood group A antigen.

BACKGROUND: Loss of blood group ABO antigens on red blood cells (RBCs) is well known in patients with leukemias, and such decreased ABO expression has been reported to be strongly associated with hypermethylation of the ABO promoter. We investigated the underlying mechanism responsible for A-antigen reduction on RBCs in a patient with myelodysplastic syndrome. STUDY DESIGN AND METHODS: Genetic analysis of ABO was performed by PCR and sequencing using peripheral blood. RT-PCR were carried out using cDNA prepared from total bone marrow (BM) cells. Bisulfite genomic sequencing was performed using genomic DNA from BM cells. Screening of somatic mutations was carried out using a targeted sequencing panel with genomic DNA from BM cells, followed by transient transfection assays. RESULTS: Genetic analysis of ABO did not reveal any mutation in coding regions, splice sites, or regulatory regions. RT-PCR demonstrated reduction of A-transcripts when the patient's RBCs were not agglutinated by anti-A antibody and did not indicate any significant increase of alternative splicing products in the patient relative to the control. DNA methylation of the ABO promoter was not obvious in erythroid cells. Targeted sequencing identified somatic mutations in ASXL1, EZH2, RUNX1, and WT1. Experiments involving transient transfection into K562 cells showed that the expression of ABO was decreased by expression of the mutated RUNX1. CONCLUSION: Because the RUNX1 mutation encoded an abnormally elongated protein without a transactivation domain which could act as dominant negative inhibitor, this frame-shift mutation in RUNX1 may be a genetic candidate contributing to A-antigen loss on RBCs.

Anemic Disease of the Newborn With Little Increase in Hemolysis and Erythropoiesis Due to Maternal Anti-Jra: A Case Study and Review of the Literature.

The severity of the hemolytic disease of the fetus and newborn (HDFN) due to Jra mismatch ranges from no symptoms to severe anemia that requires intrauterine and exchange transfusions. We encountered a newborn, born to a healthy mother having anti-Jra at 38 weeks of pregnancy, who had moderate anemia, a positive direct antiglobulin test (DAT) result, no increased erythropoiesis, and no jaundice at birth. Flow cytometry revealed that the Jra antigen of red cells in the infant was nearly negative at birth, biphasic at 5 weeks, and lowly expressed at 7 months of life. We searched online for previous case reports on HDFN due to Jra incompatibility. Among 63 reported cases, excluding 25 cases, 38 were included with the present case for analysis. Of 39 newborns, 10 developed clear anemia (hemoglobin <10.0 g/dL), and 1 died, 5 developed hydrops fetalis, 4 needed intrauterine transfusion and/or exchange transfusion, and 3 received red cell transfusion after birth; overlaps were included. Among 29 neonates with no anemia, 8 needed interventions including phototherapy and γ-globulin infusion, and the remaining 21 received conservative supports only. The maternal anti-Jra titer, ranging between 4 and 2048, did not correlate with the severity of anemia, levels of bilirubin, or any interventions required. The DAT of red cells was positive in 29 of 36 fetuses/newborns tested, whereas it was often negative among anemic neonates (4 of 9) (P < .05). Hematopoiesis did not increase effectively, as indicated by reticulocyte ratios between 1.7% and 22.3%, even with the increase in reticulocytes in anemic neonates compared with nonanemic neonates (P < .05). Total bilirubin levels ranged broadly between 0.2 and 14.3 mg/dL but were generally low. The maternal anti-Jra titer and IgG3 subclass did not correlate with the morbidity of the newborns. Being identical/compatible between mothers and their infants may possibly enhance infants' morbidity, as a weak tendency was observed (P = .053). Maternal anti-Jra may suppress erythropoiesis in fetuses via a mechanism different from the established HDFN, such as anti-D, as evidenced by the lower reticulocyte count and small increase in bilirubin in neonates. As the anti-Jra titer, IgG subclass, and DAT were not correlated with the severity, the mechanism of anti-Jra-induced HDFN remains to be elucidated.

Integrative genome analysis identified the KANNO blood group antigen as prion protein.

BACKGROUND: Anti-KANNO, a broadly reactive RBC alloantibody, is found among some Japanese pregnant women, but the genetic basis of the corresponding antigen remains unclear. STUDY DESIGN AND METHODS: We integrated a statistical approach to identify the coding gene for KANNO antigen by conducting a genome-wide association study (GWAS) on four KANNO-negative individuals and 415 healthy Japanese. We also applied whole-exome sequencing to them and performed a replication study to confirm the identified genome variation using independent 14 KANNO-negative individuals. A monoclonal antibody-specific immobilization of erythrocyte antigens (MAIEA) assay was used to locate KANNO antigen on RBC-specific membrane protein. In vivo and in vitro binding assays of anti-KANNO were further applied to the cells expressing a candidate protein. RESULTS: The GWAS revealed a genome-wide significant association of chromosome 20p13 locus (p = 2.76E-08; odds ratio > 1000 [95% confidence interval = 48-23,674]). The identified single-nucleotide polymorphism located in an intronic region of the prion protein (PRNP) gene. Whole-exome sequencing revealed a missense variant in the PRNP gene (rs1800014, E219K), which is in linkage disequilibrium with the single-nucleotide polymorphism identified in the GWAS. All 18 KANNO-negative individuals possessed the homozygous genotype of the missense variant. The MAIEA assay using anti-KANNO and mouse antihuman prion protein showed a clear difference between KANNO-positive and KANNO-negative RBCs. Anti-KANNO showed direct binding to CHO-K1 cells expressing wild-type PRNP but not to those expressing E219K PRNP. CONCLUSION: We first identified the coding gene of the high-frequency antigen KANNO located in PRNP and the missense variation (E219K) that affects the seropositivity of the KANNO antigen, which were confirmed by PRNP overexpressed cells.

Rhnull phenotype caused by a novel RHAG mutation, c.945+1G>A, in the Japanese population.

BACKGROUND: The Rh complex contributes to cell membrane structural integrity of erythrocytes. Rhnull syndrome is characterized by the absence of the Rh antigen on the erythrocyte membrane, resulting in chronic hemolytic anemia. We recently came across 3 Rhnull phenotype probands within two families with the same novel RHAG mutation in the Japanese population. MATERIALS AND METHODS: Detailed Rh phenotyping by hemagglutination was performed using monoclonal and polyclonal anti-D, -C, -c, -E, and -e; monoclonal and polyclonal anti-Rh17 antibodies; and polyclonal anti-Rh29 antibodies. RHAG mRNA transcripts were analyzed by reverse transcription-polymerase chain reaction, and the mutation was verified by genomic sequencing. RESULTS: The genomic region spanning exon 6 contained a G > A transition in the invariant GT motif of the 5' donor splice-site of Intron 6 (c.945+1G>A). The Rhnull phenotype was caused by an autosomal recessive mutation in Probands 1 and 2, determined by family history. Regarding clinical features, the degree of hemolysis varied slightly between these individuals, with Proband 3 displaying acute hemolytic anemia with an infection. While no standard therapy has been established, the condition of the patient in this study improved with conservative treatment, including hydration and antibiotics. CONCLUSION: The mechanisms of hemolysis due to the Rhnull phenotype can vary, but our findings indicate that acute hemolytic crisis caused by the Rhnull syndrome could be associated with infection.

Application of immortalized human erythroid progenitor cell line in serologic tests to detect red blood cell alloantibodies.

BACKGROUND: Antibody screening in pretransfusion tests is necessary to avoid critical complications of blood transfusion. Although red blood cells (RBCs) expressing relevant alloantigen(s) have been used for serologic antibody screening, little attention has been given to the use of cell lines, in which blood group antigen gene(s) are transduced, as reagent RBCs for antibody screening. STUDY DESIGN AND METHODS: The use of an erythroid progenitor cell line for serologic tests was studied. The expression of blood group antigens of erythroid progenitor cells was analyzed by genotyping and flow cytometry. Serologic analysis including hemagglutination was performed using erythroid progenitor cells to evaluate their sensitivity for antibody detection. Overexpression of exogenous erythroid antigen by lentiviral transduction was carried out and investigated for antibody detection sensitivity. RESULTS: Erythroid progenitor cells contained a substantial amount of hemoglobin and expressed sufficient levels of blood group antigens to detect corresponding monoclonal antibodies. Furthermore, the cell line could acquire an exogenous RBC antigen after lentiviral transduction and detected corresponding monoclonal and alloantibodies with equal sensitivity to antigen-positive RBCs. CONCLUSION: Application of erythroid progenitor cell lines for screening for unexpected antibodies could be helpful in solving issues such as reagent availability associated with the conventional RBC-based assay. The genetic expandability of erythroid progenitor cell lines by gene modification techniques could lead to the development of more convenient reagent RBCs.

Mutations of the KLF1 gene detected in Japanese with the In(Lu) phenotype.

BACKGROUND: In(Lu) is characterized by a reduced expression of antigens in the Lutheran blood group system as well as other blood group antigens. Mutations of the erythroid transcription factor, KLF1, have been reported to cause the In(Lu) phenotype, and we investigated Japanese In(Lu) to estimate the prevalence of the phenotype and KLF1 polymorphism. STUDY DESIGN AND METHODS: Blood samples were screened by monoclonal anti-CD44 and the In(Lu) phenotype was confirmed by tube tests including adsorption and elution tests using anti-Lua and anti-Lub . KLF1, LU, and A4GALT genes were analyzed by polymerase chain reaction and sequencing. RESULTS: We identified 100 of 481,322 blood donors (0.02%), and the previously characterized 20 donors, who had the In(Lu) phenotype with the LUB/LUB genotype. A total of 100 of the 120 In(Lu) individuals had mutant KLF1 alleles, and we identified 13 known and 21 novel alleles. The mutant KLF1 alleles with c.947G>A (p.Cys316Tyr), c.862A>G (p.Lys288Glu), or c.968C>G (p.Ser323Trp) were major in the In(Lu) individuals. The P1 antigen of 29 In(Lu) (two P1 /P1 , 27 P1 /P2 ) showed significantly weakened expression by hemagglutination. CONCLUSIONS: The prevalence of the In(Lu) phenotype in the Japanese population was 0.02%, and we identified 13 known and 21 novel KLF1 alleles. The KLF1 mutations cause the reduced expression of the P1 antigen.

A new ABCG2 null allele with a 27-kb deletion including the promoter region causing the Jr(a-) phenotype.

BACKGROUND: The high-prevalence antigen Jr(a) is carried on the ATP-binding cassette transporter ABCG2. The ABCG2 gene consists of 16 exons and its translation start codon is located on the second exon. Although the occurrence of the Jr(a-) phenotype is rare, several ABCG2 null alleles have been reported. We report a new ABCG2 null allele having a large deletion in this study. STUDY DESIGN AND METHODS: The Jr(a) status was determined by standard serologic tests and genomic DNA was isolated from whole blood. Exons 1 to 16 and the 5'-untranslated region of the ABCG2 gene were analyzed by polymerase chain reaction and sequencing. Expression of the ABCG2 protein on red blood cells was examined by immunoblotting. RESULTS: A Jr(a-) blood donor had a novel allele having a 27-kb deletion including noncoding Exon 1 and the promoter region of ABCG2, and the donor was apparently homozygous for the allele. In addition, we found three more individuals having heterozygosity for the same allele, with ABCG2*01N.01 having c.376C>T (p.Q126X), but did not find the allele having the 27-kb deletion in 3000 Jr(a+) individuals. Immunoblotting revealed that the ABCG2 protein was not found to be expressed in the individual with homozygosity for the ABCG2 27-kb deleted and in two individuals with an ABCG2 27-kb deleted/ABCG2*01N.01 genotype, which indirectly allows to conclude that the 27-kb deletion is responsible for a null ABCG2 allele. CONCLUSION: We first identified an ABCG2 null allele (provisional ISBT allele number ABCG2*01N.23) having a large deletion including the promoter region.

Anti-KANNO: a novel alloantibody against a red cell antigen of high frequency.

We encountered a broadly reactive red cell alloantibody in 1991, reacting unlike any other known antibody, and named it anti-KANNO after the first patient. A total of 28 cases of anti-KANNO in the Japanese literature were reviewed. To distinguish KANNO from other antibodies against high-frequency antigens, including anti-JMH, anti-Ch/Rg, and anti-Jr(a), we conducted serologic studies with proteolytic enzyme and chemical treatments, complement sensitization against red cells, and serum neutralization techniques. Reactivity of anti-KANNO against red cells lacking high-frequency antigens and antisera to high-frequency antigens against KANNO cells were tested. Among the 28 patients, 26 were female, of whom 25 had a history of pregnancy. Red cells from patient KANNO were reactive with antisera against antigens of high frequency. Anti-KANNO reacted weakly with all cells known to lack high-frequency antigens. It reacted with 2-aminoethylisothiouronium bromide, so it can be distinguished from anti-JMH. Differences among anti-KANNO, anti-Ch/Rg, and anti-Jr(a) emerged with enzyme-treated cells, complement-sensitized cells, and the addition of normal serum. As yet, there are no reports of hemolytic transfusion reaction or hemolytic disease of the fetus and newborn attributable to anti-KANNO. It appears that anti-KANNO is a newly characterized antibody more likely stimulated by pregnancy than by transfusion and with little or no clinical significance. Further surveillance and investigation of anti-KANNO, its antigen biochemistry, and its genetics are warranted.