Mitsunobu Tanaka

The RHD gene is highly detectable in RhD-negative Japanese donors.

Recent molecular studies on the Rh blood group system have shown that the Rh locus of each haploid RhD-positive chromosome is composed of two structural genes: RHD and RHCE, whereas the locus is made of a single gene (RHCE) on each haploid RhD-negative chromosome. We analyzed the presence or absence of the RHD gene in 130 Japanese RhD-negative donors using the PCR method. The RhD-negative phenotypes consisted of 34 ccEe, 27 ccee, 17 ccEE, 26 Ccee, 19 CcEe, 1 CcEE, and 6 CCee. Among them, 36 (27.7%) donors demonstrated the presence of the RHD gene. Others showed gross or partial deletions of the RHD gene. These results were confirmed by Southern blot analysis. Additionally, the RHD gene detected in the RhD-negative donors seemed to be intact through sequencing of the RhD polypeptide cDNA and the promoter region of RHD gene. The phenotypes of these donors with the RHD gene were CC or Cc, but not cc. It suggested that there is some relationship between the RHD gene and the RhC phenotypes in RhD-negative individuals. In Caucasian RhD-negative individuals, the RHD gene has not been found outside of the report of Hyland et al. (Hyland, C.A., L.C. Wolter, and A. Saul. 1994. Blood. 84:321-324). The discrepant data on the RHD gene in RhD-negative donors between Japanese and Caucasians appear to be derived from the difference of the frequency of RhD-negative and RhC-positive phenotypes. Careful attention is necessary for clinicians in applying RhD genotyping to clinical medicine.

ABCB6 is dispensable for erythropoiesis and specifies the new blood group system Langereis.

The human ATP-binding cassette (ABC) transporter ABCB6 has been described as a mitochondrial porphyrin transporter essential for heme biosynthesis, but it is also suspected to contribute to anticancer drug resistance, as do other ABC transporters located at the plasma membrane. We identified ABCB6 as the genetic basis of the Lan blood group antigen expressed on red blood cells but also at the plasma membrane of hepatocellular carcinoma (HCC) cells, and we established that ABCB6 encodes a new blood group system (Langereis, Lan). Targeted sequencing of ABCB6 in 12 unrelated individuals of the Lan(−) blood type identified 10 different ABCB6 null mutations. This is the first report of deficient alleles of this human ABC transporter gene. Of note, Lan(−) (ABCB6−/−) individuals do not suffer any clinical consequences, although their deficiency in ABCB6 may place them at risk when determining drug dosage.

High-resolution melting analysis for genotyping Duffy, Kidd and Diego blood group antigens

High-resolution melting (HRM) analysis is a simpler genotyping method than allele-specific PCR, PCR-restriction fragment length polymorphism and multiplex PCR. Duffy, Kidd and Diego are clinically important blood group antigens. We used a novel method to genotype these three blood group antigens. Purified genomic DNA extracts of blood samples (354 Duffy, 347 Kidd and 457 Diego) were amplified using specific amplification primers. HRM curves were obtained by HRM analysis. Results were in complete concordance with those obtained for previous phenotypes and genotypes. Nucleotide substitutions for these blood group antigens were differentiated by the HRM curves. HRM analysis is a simple genotyping method and is an alternative to serological typing. Our results suggest that HRM analysis can also be used for genotyping blood group antigens whose allotype specificity is determined by single nucleotide substitutions.

The i phenotype and congenital cataracts among Chinese in Taiwan

To the Editor: In their article describing a transfusion reaction after the administration of incompletely deglycerolized autologous red cells. Cregan et al.’ failed to take note of our 1982 report.* We pointed out that unacceptable concentrations of glycerol present in thawed units of previously frozen red cells can be detected easily by suspending the washed cells in either the intended recipient’s serum (compatibility test) or normal saline. When the glycerol concentration remains high after washing, readily apparent hemolysis occurs in this test. Subsequent to our report, we understand that a large number of facilities discontinued osmolarity determinations in favor of the more easily performed and more clinically related procedure that we described. Two significant events were noted in the paper by Cregan et al.’ but they were essentially ignored. 1) “A crossmatch between the patient’s serum and the transfused unit was not possible, because of the complete hemolysis of transfused red cells upon dilution with saline.” 2) “The technologist. . .recalled problems with hemolysis at the initial typing of the unit in a slide test.” Both events are to be expected when red cells are not adequately deglycerolized. These occurrences confirm the conclusion in our earlier paper that it is not necessary to pcrform a relatively cumbersome osmolarity determination to detect residual glycerol concentrations that can induce intravascular hemolysis. Further, the statement by the authors, “This case of a hemolytic reaction adds to the known risks of autologous transfusion,” impugns autologous transfusion rather than, as it should, a technical oversight. HERBERT SILVER, MD Transmion Service Harrford Hospital Harrfod CT 06115-0729 JOEL UMLAS, MD Mount Auburn Hospital Cambridge, M-4 02138 JANICE ANDERSON Harrford Hospital Harrford, CT 06115

Detection of Rh23 in the partial D phenotype associated with the DVa category

The Rh blood group system, which is the most polymorphic of the blood group systems, is of major importance in transfusion medicine. The partial D phenotypes are classified into different categories according to the absence of one or more D epitopes. Most partial D phenotypes are generated from gene conversion events between the RHD and the RHCE genes. The low-frequency antigen, Rh23 (Dw), is characteristic of the DVa category.1 The DVa red cell (RBC) is also characterized as being negative when tested with monoclonal antibodies (MoAbs) to epitopes D1 and D5 in the 9epitope model.2 The molecular basis for the DVa category has been reported, in which either the entire RHD exon 5 (DVa Hus) or a portion of it (DVa Kou) was replaced by the RHCE equivalents.3 Recently, we reported the presence of a new RHD variant (DVa-like) that closely resembles the DVa category both structurally and serologically.4 In that study, we attempted to detect the expression of Rh23 on the DVa-like RBC membrane by using a polyclonal anti-Rh23 from a donor of blood group A. After the absorption test, the antibody reacted with a control DVa category RBC in an indirect antiglobin test, showing a higher titer, and was negative with normal RBCs. Rh23 was detected on the RBC membrane in three DVa and six DVa-like individuals. These results showed that the DVa-like sample was categorized as DVa category (Table 1). Various amino acid (aa) substitutions existed on the fourth loop of the Rh polypeptides in the DVa category. Detection of Rh23 on the RBC membrane is necessary for classification of the DVa category. The molecular structures of the mutated RHD genes from the DVa (Jpn) and the International Society of Blood Transfusion (ISBT) 49 categories were very similar.4,5 Both phenotypes exhibited four aa substitutions as compared to the aa sequence in the intact D polypeptide: Phe223Val, Glu233Gln, Val238Met, and Val245Leu. However, one aa difference (Ala226Pro) between these two phenotypes was found on the fourth external loop of the D polypeptide. It is known that Pro226 and Ala226 are the E/e polymorphism in the CE polypeptide. This result indicated that the hybrid RHD-CE-D genes of DVa ( Jpn) and ISBT 49 generated through a gene-conversion event were derived from the RHe and RHE genes, respectively. Avent et al.5 also described ISBT 49 as negative when tested with some epitope D8 MoAbs. We threorized that DVa (Jpn) was different from ISBT 49. A previous study3 proposed that the aa residing at position 233 was probably involved in the Rh23 epitope, because the difference in the extracellular region between the D polypeptides and those of the DVa category was only at aa 233 (Glu/Gln), while the other aa substitution at 223 (Phe/Val) was located in the intramembrane domain in all DVa RBCs. In our previous study,4 one aa substitution was detected at 233 on the fourth loop of the Rh polypeptide in S.M. and H. K. individuals. In this serologic study, the S.M. sample with the Glu233Gln substitution was serologically positive for Rh23, while the H.K. and M.I. samples with the Glu233Lys substitution were negative (Table 1). These results suggested that Rh23, which can be expressed on the RBC membrane, results from only one aa substitution (Glu233Gln) in the D polypeptide. We also described the reactivity of the partial D RBCs (in 9-epitope model) with MoAbs (Table 2). The DVa (Jpn) category was confirmed by the lack of epitopes D1 and D5 in this study. Then, the DVa (S.M.) RBCs were negative with MoAbs to epitope D1 and with one MoAb (P3X35 [Table 2]) to epitope D5. This result may indicate that the substitution Glu233Gln was associated not only with the expression of Rh23 but also with the loss of epitopes D1 and D5 in the DVa (S.M.) sample. We also showed that one substitution, Glu233Lys in new partial D phenotype, DHK, results in the lack of various RhD epitopes (Table 2). Negative reactions were observed with MoAbs to epitopes D1, D4, D5, D6/7 (a part), and D9 (a part). This information will allow further study of the epitopes of D.

Defining the Jr(a–) phenotype in the Japanese population

The Jr(a–) phenotype is rare in European and North American populations but is not so rare in Japanese and other Asian populations. Recently, two groups have established the connection between the Jr(a–) phenotype and the ATP‐binding cassette, member G2 (ABCG2) gene and concluded that ABCG2‐null alleles encode the Jr(a–) phenotype. In Japanese Red Cross Blood Centers, the Jr(a–) phenotype is found with a prevalence of 0.05% among blood donors, and we applied DNA‐based genotyping to investigate the molecular basis of the Jr(a–) phenotype in Japan, in addition to serologic typing.

Detection and quantitation of ABO RBC chimerism by a modified coil planet centrifuge method

BACKGROUND : Differential agglutination procedures and flow cytometric analysis have been used for detecting and quantitating mixed cell populations. For more than 20 years in our laboratory, a differential agglutination method using the coil planet centrifuge and polyclonal anti‐A or anti‐B has been used. However, it is now difficult to obtain polyclonal antisera, and it is unknown whether MoAbs can take the place of polyclonal antisera in the coil planet centrifuge method.

Isolation, characterization, and family study of DTI, a novel partial D phenotype affecting the fourth external loop of D polypeptides

BACKGROUND: The Rh system is the most polymorphic of the blood group systems and is of major importance in transfusion medicine. The partial D phenotypes lack one or more of the D epitopes. These variants appear to have arisen through hybrid RhD‐CE‐D or by spontaneous point mutations in RhD. The serologic findings and the molecular characterization of a novel partial D phenotype, termed DTI, are presented here.

RHC/c genotyping based on polymorphism in the promoter region of the RHCE gene

Designing of PCR tests for the RHC allele is difficult because of the high DNA sequence homology between RHC and RHD genes, which differ by only a one-nucleotide substitution at position 48 in exon 1 of the RHCE gene. We sequenced the promoter region of the RHCE gene, and compared our results with the reported sequence. Genomic DNA was prepared from blood samples collected from 656 Japanese donors. The DNA segment encompassing the promoter region and exon 1 of the RHCE gene from 30 donors was amplified by PCR and analyzed by DNA sequencing. Four nucleotide differences between RHC/c and RHD were found at positions -468, -304, -58, and -46. On the basis of the nucleotide differences at positions -468 (RHCE vs. RHD) and -292 (RHC vs. RHc), we then developed a novel polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method for RHC/c genotyping. Analysis of the genomic DNA from the 656 donors revealed that this method could discriminate RHC from RHc, irrespective of the RHD genotype, with only a few exceptions. The combination of our system and the intron 2-based PCR-RFLP method previously reported may prove to be more accurate than either of the methods alone, and therefore, useful and valuable for RHC/c genotyping.

Novel alleles of Lan- in Japanese populations.

T he Lan antigen and antibody (anti-Lan) were first reported by van der Hart and colleagues in a severe hemolytic transfusion reaction case. The Lan− phenotype is very rare worldwide and is usually identified through the presence of anti-Lan. The Lan− phenotype has been reported in the black, Caucasian, and Japanese populations. The Japanese Red Cross Blood Centers have been screening red blood cells (RBCs) for Lan− donors since the establishment of human monoclonal anti-Lan (OSK43) in 2003. Recently, Helias and coworkers reported that purified Lan antigen from RBCs using OSK43 was identified as ABCB6 by mass spectrometry. ABCB6 is located on Chromosome 2q36 and comprises 19 coding exons that span approximately 9.2 kb of genomic DNA. It encodes ABCB6, an 842-amino-acid protein, which is a multipass N-glycosylated “halftransport protein” present on RBCs. Lan− individuals were genotyped and 10 alleles of the ABCB6 with silencing changes defining Lan− phenotypes were reported. Additionally, Saison and coworkers reported that a missense nucleotide change (c.574C>T [Arg192Trp]) in ABCB6 encodes the Lan− phenotype. To date, 19 ABCB6 alleles that encode Lan− or Lan+/−, or Lan+ phenotypes have been described. We previously reported Lan− individuals in a Japanese donor with anti-Lan. Furthermore, in this study, we report the identification of 10 new alleles and 12 new nucleotide changes in Japanese Lan− donors who do not have anti-Lan. MATERIALS AND METHODS