RHD positive haplotypes in D negative Europeans

Background Blood group genotyping is increasingly utilized for prenatal diagnosis and after recent transfusions, but still lacks the specificity of serology. In whites, the presence of antigen D is predicted, if two or more properly selected RHD-specific polymorphism are detected. This prediction must fail, if an antigen D negative RHD positive allele is encountered. Excluding RHDψ and CdeS frequent only in individuals of African descent, most of these alleles are unknown and the population frequency of any such allele has not been determined. Methods We screened 8,442 antigen D negative blood donations by RHD PCR-SSP. RHD PCR positive samples were further characterized by RHD exon specific PCR-SSP or sequencing. The phenotype of the identified alleles was checked and their frequencies in Germans were determined. Results We detected 50 RHD positive samples. Fifteen samples harbored one of three new Del alleles. Thirty samples were due to 14 different D negative alleles, only 5 of which were previously known. Nine of the 14 alleles may have been generated by gene conversion in cis, for which we proposed a mechanism triggered by hairpin formation of chromosomal DNA. The cumulative population frequency of the 14 D negative alleles was 1:1,500. Five samples represented a D+/- chimera, a weak D and three partial D, which had been missed by routine serology; two recipients transfused with blood of the D+/- chimera donor became anti-D immunized. Conclusion The results of this study allowed to devise an improved RHD genotyping strategy, the false-positive rate of which was lower than 1:10,000. The number of characterized RHD positive antigen D negative and Del alleles was more than doubled and their population frequencies in Europe were defined.


Introduction
The antigen D encoded by the RHD gene is the most important blood group antigen determined by a protein. About 15% of whites are antigen D negative. Antigen D prediction by PCR was initially applied to fetus at risk for hemolytic disease of the newbom [1,2]. If serologic blood group typing cannot be performed with its usual ease, an RHD genotyping with a specificity and sensitivity com-parable to serologic methods is of practical importance. For example, the utility of blood group genotyping in patients with recent transfusions was demonstrated by several studies [3,4,5,6].
The two RH genes, RHD and RHCE are about 30,000 bp apart [7], have opposite orientation [7,8] and are homologous retaining more than 90% identity [9]. The most frequent cause for the absence of the antigen D in whites is the lack of the whole RHD gene [10] due to a deletion occurring in the Rhesus box [7]. Therefore, most methods for antigen D prediction in whites probed the presence of RHD specific polymorphism [1,2,11,12,13]. Two RHD positive antigen D negative alleles are frequent in Africans: RHDψ carries a 37 bp insertion at the intron 3/ exon 4 boundary and also harbors a stop codon [14]; Cde s is a RHD-CE-D hybrid gene [15,16,17]. In Asians, a major allele may be associated with a G314V missense mutation [18], and several other alleles may represent RHD-CE-D hybrid alleles [18,19,20].
In whites, RHD positive antigen D negative alleles were considered rare. However, the single systematic study [21] indicated a frequency of up to 22% among the rare haplotype Cde, which would render them the major cause of false-positive antigen D prediction by PCR in whites. The majority of RHD positive alleles in D negatives were reported as scattered case reports [22,23,24,25,26,27] with an often incomplete molecular work-up. The relative frequencies of these alleles and their cumulative population frequency remained unknown.
The specificity of RHD genotyping can be improved by a systematic characterization of RHD positive antigen D negative alleles. This rationale prompted us to determine the molecular causes of such alleles and their population frequencies in a random survey among European blood bank donors. We screened more than 8,000 antigen D negative blood donations by RHD PCR, including more than 700 rare Ccddee or ccddEe samples. Nine RHD-CE-D hybrid alleles, 5 other D negative and 3 D el alleles were identified. Five D positive donors missed by routine serology were uncovered. Two anti-D immunizations were traced. We established frequency estimates for RHD positive antigen D negative haplotypes in whites, which allowed us to devise an optimized RHD PCR strategy with an enhanced and defined specificity.

Population surveys
In a first survey, we investigated 1,068 samples of blood donors that were documented as antigen D negative according to routine serologic methods. To cover the whole length of the RHD gene, we tested the RHD promoter, intron 4, exon 7, and the 3' untranslated region of exon 10 by PCR-SSP ( Table 1). As antigen D negative RHD gene positive alleles are known to preferentially occur in the Cde and cdE haplotypes [21,22,25], we tested 754 samples with antigen C or antigen E or both along with 314 ccddee samples. We detected 48 donors who carried the RHD gene. All were positive for antigen C or antigen E or both (Table 1).
In a subsequent survey, we checked 7,374 ccddee samples, which were tested in pools of twenty samples for RHD promoter, intron 4 and exon 10. This survey aimed to increase the power of our study for ccddee donors, which represent 92% of all antigen D negative [28]. Two RHD positive donors were detected (Table 1). In summary, 50 RHD positive donors were found in the two population surveys. They were further characterized by a detailed molecular work-up including RHD exon specific PCR-SSP, PCR of intron polymorphism or nucleotide sequencing.  1B) and were characterized by one missense and two splice site mutations, respectively. For each allele, only a single sample was sequenced, and the influence of the mutations on mRNA splicing was not verified by cDNA analysis. Because lack of material, we could not formally exclude the D el phenotype for the two RHD-CE(2-7)-D 1 samples (Fig. 1A). However, a D el expressed by this allele was very unlikely, as several other hybrid alleles carrying smaller gene conversions, like RHD-CE(2-7)-D 2 , were unequivocally D negative.

Population frequencies
The population frequencies of the alleles were calculated ( Table 2). The cumulative frequency of all antigen D negative RHD gene positive haplotypes was estimated to be 1:1,537. The most frequent allele was RHD-CE(2-9)-D 2 with a frequency of 1:5,682, representing about 27% of antigen D negative RHD gene positive alleles. Hybrid alleles lacking RHD exon 4 to exon 7 accounted for 68% of antigen D negative RHD gene positive alleles. 84% of antigen D negative RHD positive haplotypes carried the antigen C, compared to less than 3% of all D negative haplotypes [28]. The cumulative allele frequency of D el was 1:3,030.

Analysis of SCARF samples
We obtained 2 ccddEE, 1 CcddEe and 1 ccddee (G+) DNA samples through the SCARF Exchange and tested them for promoter, intron 4, exon 7 and the 3' untranslated region by RHD PCR-SSP. Positive reactions were obtained with the CcddEe sample only, which was assigned to the RHD-CE(8-9)-D allele by RHD exon-specific PCR-SSP.

Optimized RHD PCR
Based on the population frequencies, we calculated the expected positive predictive values of a positive result for different RHD PCR strategies (Table 3). RHD PCR based on intron 4 and exon 7 had a considerably higher positive predictive value than testing exon 10 alone. Testing for RHDψ [14] improved specificity. Even greater improvements were effected by testing for other allele like RHD-CE(8-9)-D or RHD(W16X). An optimized PCR strategy would comprise checking for RHD intron 4, exon 7 and intron 7 complemented by the specific detection of RHD(W16X) and RHDψ Antigen prediction in the rare samples positive for either of these alleles necessitates complementary methods, since the allele in trans may be D positive. The five polymorphism tested can be multiplexed in two PCR tubes (Fig. 5). This assay was about twice as reliable as current exon scanning approaches [6,26] requiring, if not multiplexed [26], up to 8 separate PCR tubes [6].

Gene conversions in cis
The gene conversion of five hybrid alleles observed in this study involved intron 2 ( Fig. 1) which was utilized to delineate the allele origin of the RHCE gene segment found in the RHD gene. In all five haplotypes, a gene conversion in cis was likely (Fig. 6A). We proposed that gene conversions in cis occur during hairpin formation, which is favored by the clustered gene arrangement (Fig. 6B).

Anti-D immunizations
A flow cytometry study of the RBC from the D +/chimera revealed 94% D negative RBC and an admixture of 6 % D positive RBC (Fig. 7). This chimera was confirmed in a 3 month follow-up. The 24 year old donor was healthy and had no twin. A look-back of this donor revealed 13 units that had been issued as D negative. Two D negative recipients were traced and available for an antibody screen, both of whom were anti-D immunized.

Discussion
In a systematic population survey including more than 8,000 antigen D negative blood donations, we identified 14 different RHD positive antigen D negative and 3 different D el haplotypes, the majority of which were novel. The molecular bases were alleles comprising RHD/ RHCE hybrids, stop codons, missense mutations and splice site mutations. The cumulative frequency of RHD gene positive antigen D negative haplotypes was about 1:1,500; that of the D el alleles was about 1:3,000. We determined the specificity of antigen D prediction by PCR and devised an optimized RHD PCR strategy with a calculated positive predictive value greater than 0.9999. Five antigen D positive samples missed by routine D typing were uncovered and two anti-D immunizations traced.
For practical purposes, two groups of RHD alleles that do not express antigen D can be distinguished. RHD alleles of the first group lack some or many RHD specific polymorphism and usually represent RHD/CE hybrids. For alleles of this group, a correct antigen D prediction may be accomplished by a prudent selection of the RHD specific polymorphism utilized for RHD genotyping. RHD alleles of the second group carry all RHD specific polymorphism and most often harbor point mutations. For alleles of this group, a correct antigen D prediction necessitates the specific detection of an aberration that is usually unique to the allele. The identification of four new alleles in this group increased the number of known alleles from 3 to 7 and was critical for improving RHD genotyping.
The data of this study allowed for the first time to calculate population frequencies of RHD positive antigen D negative and D el alleles. This information was indispensable to derive rational RHD typing strategies and will be essential for establishing cost-efficient approaches. The majority of samples belonging to the first group of D negative alleles (probable RHD/CE hybrids) was compatible with RHD-CE-D hybrid alleles, in which the DNA segment derived from the RHCE gene encompassed at least exon 4 to exon 7. These samples would be correctly typed, if exon 4/intron 4 and exon 7 were used for RHD genotyping, as proposed previously [13]. With the exception of RHD exon 9, testing additional RHD exons would not have improved the specificity of antigen D prediction. Improving this specificity, however, became possible by the specific detection of frequent alleles of the second group, like RHDψ and RHD(W16X). We demonstrated that testing 5 carefully selected polymorphism would have resulted in an assay yielding false positive results at a rate less than 1:12,000, and hence would have doubled the specificity compared to contemporary approaches testing all informative RHD exons [6,26]. Further improvements may be achieved by the specific detection of additional alleles, that might become practical in massively parallel molecular assays.
The detailed analysis including intron polymorphism revealed that the first group of alleles (probable RHD/CE hybrids) represented at least 9 different molecular events. We proposed that the proximity and inverse orientation of both RH genes favored gene conversions occurring in cis (Fig. 6), which have also been noted in partial D [31]. An exact definition of the molecular bases of the RHD/CE hybrids would allow their specific detec-  However, checking D-negative samples, especially those occurring with a C or E or both, for RHD specific sequences by nucleic amplification techniques may become practical in the near future. The knowledge of the detected alleles is also important for fetal genotyping assays using fetal DNA in maternal plasma, because false positive results will be obtained in mothers harboring RHD positive D negative alleles.

Blood samples
EDTA-or citrate-anticoagulated blood samples were collected from blood donors characterized as D negative in routine typing including an antiglobulin test with anti-D.
The D antigen determination in antiglobulin technique was performed as part of routine blood donor typing over a period of more than 15 years with varying commercial anti-D in tube or column agglutination. For each donor, this antiglobulin test was done only once, if the donor lacked the antigens C or E, and two times from independent samples, if the donor was either C or E positive. Subsequent donations were checked by direct agglutination using an Olympus PK7200 autoanalyzer only. Samples were collected at random for specific CcEe phenotypes. DNA was isolated by a modified salting-out procedure as described [27,39].

Screening by PCR with sequence specific priming (PCR-SSP)
For the first population survey, 314 ccddee, 433 Ccddee, 271 ccddEe, 19 CcddEe, 24 CCddee, 1 CcddEE and 6 ccd-dEE samples were tested individually for the presence of RHD specific polymorphism located in the RHD promoter, intron 4, exon 7 and the 3' untranslated region of exon 10 by PCR-SSP. The donor previously reported as "CCDnex ee" [27] returned by chance and his allele was further characterized as RHD-CE(8-9)-D.
For the second population survey, 7,374 ccddee samples were analyzed in pools. Equal volumes of 20 samples were mixed. To confirm the sensitivity of the pool testing, 1% of D positive blood was added to an aliquot as positive control. DNA was extracted and checked for RHD promoter, intron 4, and exon 10 using modified PCR-SSP to enhance sensitivity. Repeated testing of donors was minimal, because the mandatory donation interval exceeded the collection period.
The donors were representative for the population currently living in the county of (Land) Baden-Württemberg. The ethnic origin of individual donors, in particular of those carrying Cde s or RHDψ was not identified. Independent of the population surveys, 2 ccddEE, 1 CcddEe and 1 ccddee DNA samples of unknown ethnic backgrounds were obtained from the SCARF Exchange (Hahnemann University, Philadelphia, USA).

Further molecular characterization
All samples positive for any of the above mentioned PCR-SSP assays were further investigated for the presence of RHD specific polymorphism in exon 3, exon 4,

Figure 1
Predicted molecular structure of the 17 RHD positive, D negative or D el alleles detected. For each haplotype, a schematic representation of the molecular structure is shown along with a designation, haplotype association, phenotype, and numbers of samples observed. Each RHD exon is indicated by a box, intron and promoter polymorphism investigated are shown as circles.
White symbols indicate the presence of RHD specific sequences, black symbols their lack as predicted form the RHD exon-specific PCR-SSP results.

RHD-CE(3-7)-D
A B exon 5, exon 6, exon 7, and exon 9 by PCR-SSP. Samples positive for all PCR-SSP were sequenced, until they could be assigned to a distinct RHD allele. Samples negative for some PCR-SSP were checked for informative polymorphism in intron 1, intron 2, intron 7, intron 8, and intron 9.

Analysis for Cde s and RHDψ RHD(N152T) and RHCE(L245V) present in Cde s [16]
were checked by PCR-SSP. The 37 bp insertion present in RHDψ was detected by PCR-SSP. The 37 bp insertion, the M218I, F223V and S225F missense mutations and the Y269X nonsense mutation previously described for RHDψ [14] were confirmed by sequencing of all 10 exons; no additional aberrations were detected.

Nucleotide sequencing
The ten RHD exons were sequenced as described [29,32]. The promoter was amplified with primers rend31k (for RHD alleles) or re04 (for RHCE) and rb45 and a DNA stretch encompassing primer re012 was sequenced using primers re08 and re09.
Intron 1 was tested 1173 and 1174 bp 5' of the intron 1/ exon 2 boundary by RHD specific amplification of exon 2 as described [32]. Intron 2 was evaluated by PCR with length polymorphism as described [31]. The BamHI restriction site introduced by the 9 bp deletion in RHD intron 8 position 1114 to 1122 (Genbank accession number AL139426) was checked after amplification with primers re74 and re93 and digestion with BamHI. The 980 bp deletion starting at position 633 in RHD intron 9 (Genbank accession number AL139426) was evaluated using primers re93k and re916.

Flow cytometry
Flow cytometry was performed as described [40,41] using a polyclonal anti-D (anti-D Molter; Ortho Clinical Diagnostics, Neckarsgmünd, Germany) as primary and goat anti-human IgG, F ab -fragment, FITC-conjugated  Markers were set to encompass >99.5% of a D positive control and less than 0.5% of a D negative control. The percentage of cells in the marker area was evaluated.

Haplotype frequencies
For alleles observed more than once, the haplotype association with Cde and cdE was obvious, because of their repeated observations in association with the rare phenotypes Ccddee or ccddEe, respectively. Based on the paucity of RHD positive samples among the ccddee samples, alleles that were observed only once were assumed to be associated with the Cde or cdE haplotype rather than the cde haplotype. An allele occurring in an unique CcddEe sample was counted as Cde. The RHDψ allele was assumed to be associated with the ce(W16C) allele, because RHCE specific sequencing of exon 1 revealed a C/G heterozygosity at position 64 and ce(W16C) is almost absent from the cde haplotypes in our population [27]. The frequency of a given aberrant RHD allele in its haplotype was calculated as the number of observed samples divided by the number of the corresponding haplotypes under observation (500 Cde, 303 cdE). For cde, the haplotype frequency was calculated from the 14,748 haplotypes checked in the second survey. The population frequency of an RHD allele was calculated from the frequency of this allele in its haplotype and the known frequency of the haplotype in the local population  The internal control amplicons, which were devised to be larger than the specific amplicons, may be suppressed because of competition, if a specific product is amplified.

Figure 6
Gene conversion in cis. Panel A: Origin of the RHCE gene segments. The allele origin of the RHCE segments in the RHD gene was analyzed by a PCR length polymorphism in intron 2 [31,43]. The 1,177 bp product is specific for the C allele of RHCE, the 1,068 bp product for the c allele of RHCE and for RHD. The CcDee control shows a strong band at the c/D position and a weaker band at the C position (lane 1). The cE associated RHCE(1-9)-D hybrid allele lacks the C band (lane 2), indicating that the intron 2 of the hybrid allele derives from c. In contrast, all Ce associated hybrid alleles involving intron 2 show a strong C band and a weaker c/D band (lanes 3 to 6), indicating that the introns 2 of those hybrids derive from C. Panel B: Proposed mechanism of gene conversion in cis. (i) The RHD and RHCE genes are inversely orientated [7] as typical for clustered genes. (ii) A putative hairpin formation of the chromosome allows the close proximity of homologous segments in identical orientation. This structural feature is instrumental for gene conversion events in cis. (iii) Resolving the hairpin yields an RHD-CE-D hybrid gene structure, many of which have been observed to date at the RH gene locus. As an example, the RHD-CE(4-7)-D hybrid exon structure is shown. Symbols are according to Fig. 1.