Transfusion support for a woman with RHD*09.01.02 and the novel RHD*01W.161 allele in trans

Abstract

According to recent work group recommendations, individuals with the serologic weak D phenotypes should be RHD genotyped and individuals with molecular weak D types 1, 2, 3, 4.0, or 4.1 should be treated as D+. We report an African American woman with a long-standing history of metrorrhagia, who presented for infertility evaluation. Blood grouping showed AB with a possible subgroup of A, based on mixed-field agglutination, and a serologic weak D phenotype. Results from routine red cell genotyping for the RHD gene was incongruent with the serologic RhCE phenotype. For the surgical procedure, the patient was hence scheduled to receive group AB, D– RBC transfusions. Subsequent molecular analysis identified the ABO*A2.01 and ABO*B.01 alleles for the ABO genotype and the novel RHD allele [NG_007494.1(RHD):c.611T>A] along with an RHD*09.01.02 allele for the RHD genotype. Using a panel of monoclonal anti-D reagents, we showed the novel RHD(I204K) allele to represent a serologic weak D phenotype, despite occurring as a compound heterozygote, designated RHD*weak D type 161 (RHD*01W.161). Individuals with a weak D type 4.2 allele are prone to anti-D immunization, while the immunization potential of novel RHD alleles is difficult to predict. For now, patients should be treated as D– in transfusion and pregnancy management, when they harbor a novel RHD allele along with any weak D allele other than weak D types 1, 2, 3, 4.0, or 4.1. This study exemplifies strategies for how and when a laboratory should proceed from routine genotyping to nucleotide sequencing before any decisions on transfusion practice is made.

RhD-mediated hemolytic disease of the fetus and newborn (HDFN), although of much lower incidence than in the 1960s, continues to pose a serious risk for the fetus or neonate.^1–5 HDFN caused by anti-D can also occur, when mothers carry variant RhD proteins.^6–11 Recent work group recommendations advocate the use of RHD genotyping to resolve serologic weak D phenotypes.^12,13 They also recommend individuals carrying the molecular weak D types 1, 2, 3, 4.0, or 4.1 to be considered as D+.¹³ However, female patients hemizygous for any of the many weak D alleles^14–16 other than weak D types 1, 2, 3, 4.0, or 4.1 or novel RHD alleles should still be treated as D– for transfusion and pregnancy management.¹⁷

The clinically relevant DAR1.2 (weak D 4.2.2) phenotype encoded by the RHD*09.01.02 allele,¹⁸ initially designated as weak D type 4.2.2,¹⁹ belongs to the weak D type 4 cluster²⁰ and is prevalent among individuals of African ancestry.¹⁹ It is characterized by three non-synonymous [602C>G (T201R), 667T>G (F223V), 1025T>C (I342T)] and two synonymous [744C>T (S248S), 957G>A (V319V)] variations in the RHD gene and is associated with the cDe haplotype.^19,21 Carriers of the RHD*09.01.02 allele are at risk for anti-D alloimmunization, because the encoded D antigen lacks some epitopes.^19,22

The weak D type 19 phenotype encoded by RHD*01W.19 allele,¹⁸ closely related to the novel allele described in this study, has only been observed once in a white blood donor.²³ It is characterized by the single non-synonymous variation c.611T>C (I204T) in exon 4 of the RHD gene and reported to be associated with the CDe haplotype.²³ With 2456 D antigens per red blood cell (RBC) (normal range for CcDe phenotype is 6500–17,500 antigens/RBC), the weak D type 19 still presents itself as a D antigen of normal strength using standard serologic methods.²³ There are no reports of anti-D immunization in weak D type 19 individuals since its description in 2006.²³

We report a woman with a serologic weak D phenotype harboring a novel allele in trans to a clinically relevant known RHD allele. The presence of two different variant alleles at a particular gene locus is known as compound heterozygosity.²⁴ The workup in this case demonstrated the clinical utility of resolving a serologic weak D phenotype.¹² Our study also illustrated weakened A antigen expression by RBCs harboring an ABO*A2.01 allele with an ABO*B.01 allele in trans and mistyping of RHD alleles using commercial genotyping assays. We discuss the possible mutual influence of two different weak D alleles on the RhD phenotype in compound heterozygous individuals and the impact of different amino acid substitutions on the three-dimensional structure of the RhD protein and D-antigen density.

Case Report

A 30-year-old African American female patient (gravida 0) presented for infertility treatment. She had polycystic ovary syndrome, endometriosis, galactorrhea, prolactinoma, dysmenorrhea, and a long-standing history of metrorrhagia. She had previously received 3 group A, D+ RBC units due to symptomatic anemia with syncope. At an outside hospital, a robotic-assisted metroplasty to correct uterus didelphys was performed, at which time she was typed as group AB, D+. She was enrolled 9 months later in the National Institutes of Health clinical protocol 99-CH-0103 for “Evaluation of Women and Men with Endocrine and Reproductive-Related Conditions” and scheduled for a diagnostic laparoscopy, chromopertubation and endometrial biopsy, and dilation and curettage followed by operative hysteroscopy.

Blood grouping showed AB with a possible subgroup of A, a serologic weak D phenotype, and a negative antibody detection test. In forward grouping, the agglutination strength of A antigen was 3+, weaker than B (4+), but showed a mixed-field agglutination pattern. Anti-A₁ lectin testing was negative, as was the reverse grouping with A₁, B, and A₂ reagent RBCs. For routine care, the patient was noted to receive group AB RBCs.

The patient showed no agglutination with anti-D at immediate spin via tube method but showed 3+ agglutination via the gel matrix method. Weak D testing with antihuman globulin showed 2+ agglutination (tube). Thus, testing reproducibly confirmed a serologic weak D phenotype (Table 1). Following our standard operating procedure, molecular screening of the patient’s RHD gene was performed using the RBC-FluoGene D weak/variant kit, which predicted an RHD*01W.14 allele. The lack of an RHCE*cE allele, however, which is typically associated with the RHD*01W.14 allele, prompted us to sequence the RHD gene. Nucleotide sequencing of the 10 RHD exons identified all nucleotide variations associated with RHD*09.01.02 along with the additional variation c.611T>A. Because of the likely presence of RHD*09.01.02, the patient was to be managed as group AB D– for the planned surgery.

Table 1.

Routine serology

Test	Result
ABO group	AB (mixed-field agglutination for A)*
RhD phenotype	Serologic weak D phenotype†
RhCE phenotype	C+E–c+e+
Antibody detection test	Negative
Direct antiglobulin test	Negative

^*Anti-A₁ negative. Molecular characterization consistent with A₂B phenotype.

^†D typing in tube agglutination using a routine oligoclonal anti-D blend (BS232, BS221, and H41 11B7; Bio-Rad, Hercules, CA) was negative in immediate spin but 2+ positive with antihuman globulin (AHG). D typing in gel matrix method (clone MS-201) with AHG was 3+ positive.

The procedure was uneventful, without the need for blood transfusion. Subsequently, an extended molecular workup was performed that documented an A₂B subgroup and confirmed a suspected novel RHD allele along with an RHD*09.01.02 allele, thus asserting our initial recommendation to manage the patient as group AB, D–.

Materials and Methods

Immediate spin test and indirect antiglobulin test were performed as per our standard operating procedure by standard tube and gel matrix methods with licensed reagents (Ortho Clinical Diagnostics, Raritan, NJ, and Bio-Rad Laboratories, Feldkirchen, Germany) and 13 monoclonal anti-D reagents (Advanced Partial RhD Typing Kit and RhD Variant Investigation Kit; Alba Bioscience/Quotient, Eysins, Switzerland).

To characterize the ABO allele responsible for the possible A subgroup, we performed nucleotide sequence analysis of all seven exons of the ABO gene as described previously.²⁵

To characterize the RHD allele responsible for the serologic weak D phenotype, we analyzed the patient’s RHD gene using three different molecular methods. TaqMan Probe–based analysis was performed using the RBC-FluoGene D weak/variant kit (Inno-Train Diagnostik, Hesse, Germany), DNA array analysis using the wRHD BeadChip from BioArray Solutions (Immucor, Warren, NJ), and nucleotide sequence analysis of all 10 exons of the RHD gene using Sanger sequencing as described previously.^26–28 Zygosity for the RHD gene was determined by restriction fragment-length polymorphism.²⁹

Because of lack of fresh whole blood for mRNA isolation, allele-specific amplification of genomic DNA and sequencing was performed to separate and identify the two RHD alleles. The following primers were designed and used for amplification and sequencing:

• rf4-6ds: TAAGCTCTGAACACCAGTCTCA (forward)²⁷ located in RHD intron 3,

• RHD_3RC: CCTGAGATGGCTGTCACCAAG (reverse), and

• RHD_3RT: CCTGAGATGGCTGTCACCAAA (reverse).

These primers targeted the wild-type and variant nucleotides of 744C>T substitution (underlined) in RHD exon 5 of RHD*09.01.02 allele. Specificity of the assay was enhanced by introducing a mismatch in the penultimate 3′-nucleotide in the reverse primers (italicized).

Results

In routine serology, the patient typed as group AB with mixed-field agglutination for A. She also typed as a serologic weak D phenotype (Table 1). Her Rh phenotype was C+E– c+e+.

ABO Alleles

Sequencing of all seven exons of the ABO gene revealed the genotype of the patient to be ABO*A2.01/ABO*B.01 compatible with the group A₂B phenotype determined by serologic testing.

RHD Alleles

Using the RBC-FluoGene D weak/variant kit in routine care, the nucleotide c.602G was detected, which can indicate a weak D type 4 or a weak D type 14 or both (compound heterozygous). The nucleotide c.544T, which is present in weak D type 4 but replaced with c.544A in weak D type 14, was not detected. Hence, the software algorithm concluded indirectly that the patient carried the weak D type 14 allele (Table 2). Further, using the wRHD BeadChip, the patient typed as heterozygous with an RHD*DAR (RHD*09.01)¹⁸ and a possibly normal RHD allele. Finally, using Sanger sequencing in conjunction with newly designed allele-specific amplification, the patient was determined to be compound heterozygous with a novel RHD allele (c.611T>A) in trans to the known RHD*09.01.02 allele^16,19 (Table 2). The patient was confirmed to carry two copies of RHD gene (RHD homozygous) by restriction fragment-length polymorphism.

Table 2.

Red cell genotyping of the RHD gene

	RHD alleles
Test	Allele 1	Allele 2
RBC-FluoGene D weak/variant	RHD*01W.14	Possible RHD*01W.14 or RHD*01 or RHD*01N.01
wRHD BeadChip	RHD*DAR	Possible RHD*01
Nucleotide sequencing
Allele recognized	Known	Novel
Nucleotide sequence (GenBank)	KF712273	MT980847
Amino acid change(s)	T201R, F223V, I342T	I204K
Serologic weak D phenotype15	DAR1.2 (weak D 4.2.2)18	Weak D type 161
ISBT terminology16	RHD*09.01.0218	RHD*01W.161†

^†ClinVar SCV001977615.

Serologic Characterization

We tested the strength of the D antigen expressed on the patient’s RBCs. Except for LHM57/17, the remaining 12 monoclonal anti-D reagents³⁰ agglutinated the RBCs with varying reaction strengths (Table 3). Blood samples from individuals hemizygous for weak D type 45 (GenBank AJ867388), weak D type 64 (GenBank AM902713), and weak D type 88 (GenBank LN612637) alleles were tested in parallel with the patient’s RBCs (Table 3). For comparison, the same RBC samples were also tested in immediate spin in tube and gel matrix (Table 4).

Table 3.

Serologic testing with panels of anti-D

			D variant and serologic reaction strength*
			Study sample	Control samples
Monoclonal anti-D†				Weak D type
Clone	Isotype	epD	Patient CDe/cDe	19‡ CDe/cde	1 CDe/cde	4.0§ cDe/cde	4.1 cDe/cde	4.2.2¶ cDe/cde	45 CDe/cde	64 cDe/cde	88 cDE/cde	D+ cDe/cde	D− cde/cde
LHM76/58	IgG1λ	8.1	4+	+	4+	4+	4+	3+ to 4+	4+	4+	4+	4+	0
LHM76/59	IgG1	15.1	4+	+	4+	4+	4+	3+ to 4+	4+	4+	4+	4+	0
LHM174/102	IgG3k	1.2	2+	+	0	0	0	0	4+	2+	1 +	4+	0
LHM50/2B	IgG1λ	6.3	4+	NT	4+	4+	4+	3+ to 4+	4+	4+	4+	4+	0
LHM169/81	IgG3k	1.1	4+	+	4+	4+	4+	2+ to 3+	4+	4+	4+	4+	0
ESD1	IgG1k	4.1	4+	NT	4+	4+	4+	3+ to 4+	4+	4+	4+	4+	0
LHM76/55	IgG1k	3.1	4+	+	4+	4+	4+	3+ to 4+	4+	4+	4+	4+	0
LHM77/64	IgG1k	9.1	4+	NT	4+	4+	4+	3+ to 4+	4+	4+	4+	4+	0
LHM70/45	IgG1λ	1.2	3+	+	3+	3+	3+	0 to 2 +	3+	3+	3+	4+	0
LHM59/19	IgG1k	8.1	4+	NT	3+	3+	3+	0 to 2 +	4+	3+	3+	4+	0
LHM169/80	IgG3λ	6.3	4+	NT	4+	4+	4+	3+ to 4+	4+	4+	4+	4+	0
LHM57/17	IgG1λ	6.3	0	NT	0	0	0	0 to 2 +	NT	0	0	1 +	0
LDM1	IgM	6.5	3+	NT	0	4+	3+	0 to 2 +	4+	0	2+	4+	0

^*Gel matrix test with antihuman globulin.

^†ALBAclone Advanced Partial RhD typing kit (recently renamed as ALBAclone RhD Variant Investigation kit) (Alba Bioscience/Quotient, Eysins, Switzerland).

^‡Results from Yu et al.²³

^§Two samples tested.

^¶Four samples tested.

NT = not tested.

Table 4.

Immediate spin testing

Samples
	Study sample	Control samples
		Weak D type
Test method*	Patient CDe/cDe	4.0† cDe/cde	4.1 cDe/cde	4.2.2‡ cDe/cde	45 CDe/cde	64 cDe/cde	88 cDE/cde	D+ cDe/cde	D− cde/cde
Tube	0	2 +	1+	0	2+	0	0	3+	0
Gel matrix	3+	3+	3+	1+ to 3+	4+	3+	3+	4+	0

^*For the tube testing, clones BS232 (IgM, epD 6.4), BS221 (IgG, epD 6.3), and H41 11B7 (IgG, epD 3.1) were used, and for gel matrix testing, clone MS-201 (IgM, epD 6.1) was used. Per test principle, all immediate spin tests are performed without antihuman globulin.

^†Two samples tested.

^‡Four samples tested.

Nomenclature

The name weak D type 161 was assigned to the novel weak D allele (RHD blood group alleles v6.0)¹⁸ in continuation of the weak D allele numbering system established in 1998.³¹

Discussion

This clinical report illustrates the application of red cell genotyping in routine clinical care. The patient was compound heterozygous for two different variant RHD alleles. Traditional serologic methods cannot resolve the underlying genotypes in such patients. Most often, complete nucleotide sequencing of the coding region of ABO and RHD is still unnecessary. However, routine red cell genotyping fails when new alleles are encountered, as exemplified in our study.

The African American female patient presented for routine serology testing in preparation for surgery. Initial serologic workup indicated her RBCs to be group AB with mixed-field agglutination for A and a weak D phenotype due to a lack of D reactivity at immediate spin. The concurrent occurrence of variants for both ABO and Rh antigens is rare and can mimic a two-cell population, such as in chimera^32,33 and mosaicism.^34,35 Per standard operating procedure, we apply red cell genotyping for all serologic weak D phenotypes to determine the specific causal RHD alleles.

Once her molecular weak D type was confirmed, we addressed the mixed-field agglutination by ABO gene sequencing and identified an ABO*A2.01 allele³⁶ and an ABO*B.01 allele. RBCs from group A₂B individuals are known to react weakly with anti-A reagent in direct agglutination tests.³⁷ Group A₂B RBCs rarely cause mixed-field agglutination,³⁸ which is generally the hallmark of the A₃ subgroup.³⁹ None of the variants associated with the six known A₃ alleles⁴⁰ or any novel variant was identified in the ABO gene. The heterogeneity of erythrocyte antigen distribution⁴¹ along with the AB phenotype may have caused the mixed-field agglutination, not typically expected for the A₂ phenotype, and erroneously pointing to the possibility of two RBC populations.^32–35

The RHD genotyping kit applied in our routine clinical protocol incorrectly determined an RHD*01W.14 allele^26,42 (Table 2). This molecular result was discordant with the African American ethnicity of the patient and the serologic Rh phenotype CcDe, because RHD*01W.14 has previously only been observed in the white population and is associated with the RHCE*cE allele,²⁶ whereas this patient was E–. Another commercial RHD genotyping assay identified a DAR allele and predicted a possibly conventional RHD allele in trans (Table 2). Of course, the presence of a conventional RHD allele would conflict with the negative serologic reactivity for D in the immediate spin tube test (Table 4). Nucleotide sequencing of the RHD gene resolved these dicrepancies and identified a novel RHD allele, RHD*c.611T>A (p.Ile204Lys) dubbed weak D type 161 (RHD*01W.161), in addition to an RHD*09.01.02 allele (Table 2).

We routinely perform nucleotide sequencing for clinical purposes when discrepancy with published data such as RH haplotype and ethnicity is found. For scientific purposes, we also perform nucleotide sequencing to comprehensively analyze novel or less well characterized RHD alleles and to establish their long-range RH haplotype sequences.

Anti-D immunization rates can surpass 50 percent in many clinically relevant situations.⁴³ However, despite previous multiple D+ transfusions, no anti-D immunization had occurred in the current patient. The lack of such observations is relevant, particularly if documented for larger cohorts.¹³ It may be possible in the future that the vast majority of patients with most of the more than 161 molecular weak D types, including the patient presented here, can safely be treated as D+.¹⁷ Current recommendations do not support such a policy, but they may be revised based on forthcoming clinical evidence, which hence needs to be published.

The amino acid position p.204 is located in the RBC membrane near the intracellular region of the RhD protein⁴⁴ and is associated with two variant alleles. The wild-type RhD protein has a non-polar isoleucine, which is replaced with the polar but uncharged amino acid threonine (p.I204T or Ile204Thr) in the known RHD*01W.19 allele. In the novel RHD*01W.161 allele, a positively charged amino acid lysine (p.I204K or Ile204Lys) occurs at that position instead. Isoleucine and threonine are both β-branched amino acids and restrict the conformations that the main chain can adopt.⁴⁵ However, isoleucine, being a neutral amino acid, lacks hydrogen-bonding potential, whereas threonine, with a hydroxyl group in its side chain, and lysine, with a positively charged amino group in its side chain, can easily form hydrogen bonds. Thus, the variant amino acids threonine and lysine may alter the electrostatic nature of the transmembraneous α-chain and destabilize the RhD protein or alter the specificity and affinity of the RhD protein’s interaction with the underlying cytoskeletal proteins.^46–48 This mechanism may also explain the weak D phenotype caused by the nearby variation at amino acid position p.202 (p.A202V or Ala202Val) in the RHD*01W.43 allele.⁴⁹

Both weak D type 19 (p.Ile204Thr) and the DAR1 (weak D 4.2)¹⁸ (p.Thr201Arg, p.Phe223Val, p.Ile342Thr) express significantly reduced D antigen densities of 2721²³ and 1650,¹⁹ respectively. The DAR1.2 (weak D 4.2.2) lacks certain RhD epitopes and may test negative with clones LHM174/102, LHM70/45, LHM59/19, and LDM1 (Table 3).²⁸ However, these four anti-D reagents agglutinated the patient’s RBCs. Based on this reactivity pattern, we concluded that the novel weak D type 161, despite occurring in trans to a DAR1.2 (weak D 4.2.2), has a weak D phenotype instead of a DEL or a D– phenotype. The lack of reactivities with the clone LHM57/17 was possibly caused by the known instability of the clone during prolonged storage, as shown by the weak reactivity (1+) in the D+ control (Table 3).

The weak D type 19 types as a D antigen of normal strength by routine serology, while the DAR1 (weak D 4.2) is known to give disparate results in the immediate spin test as shown before⁵⁰ and also in the current study (Table 4). Weak D alleles are typically observed in isolated form in individuals known as hemizygotes with only one copy of the variant RHD gene. However, individuals may carry two different copies of RHD gene and are known as being compound heterozygotes.^51–56 The two variant RhD proteins in these individuals can interact with each other and hinder the integration of RhD protein in the RBC membrane. This phenomenon is known as dominant-negative inhibition⁵⁷; a hemizygous weak D type 161 sample would be needed to examine this possible mechanism.

The present clinical report illustrates the advantage of molecular methods to identify novel or known RHD alleles and allow the physician to make an informed decision for the patient to be safely treated as D+ or D–.¹³ Red cell genotyping should be performed as soon as possible once a serologic weak D phenotype is recognized,¹² either as an in-house test or through a reference laboratory. Such molecular screening will help in the resolution of discordant D typing results and will prevent the unnecessary use of D– RBCs, which are often in limited supply, and will also avoid the unnecessary administration of Rh immune globulin.¹³

The encounter of most rare alleles and novel alleles is expected to yield inconclusive results as exemplified in the current case study by discrepancies in RH haplotypes, serologic reaction strengths, and ethnic association. The causes for such discrepancies should be identified for the benefit of patient care and to advance allele databases. Identifying compound heterozygous individuals can be clinically relevant because patients with weak D types 1, 2, 3, 4.0, or 4.1 will not need D– RBCs or Rh immune globulin.¹³ In the current case, however, the original D– strategy was confirmed.