Download Complexities of the Dombrock blood group system

45S29299Original Article
Blackwell Science, LtdOxford, UKTRFTransfusion0041-11322005 American Association of Blood BanksAugust 2005 Supplement
DOMBROCK BLOOD GROUPREID
Complexities of the Dombrock blood group system revealed
Marion E. Reid
The Doa antigen was discovered after I began my career
in immunohematology and I have been fortunate to
be involved in several fascinating discoveries in the
Dombrock blood group system. The Doa antigen and its
antithetical antigen, Dob, have a prevalence that makes
them useful as genetic markers. The paucity of reliable
anti-Doa and anti-Dob has prevented this potential from
being realized; however, our ability to type for DO alleles
at the DNA level has made it possible to test cohorts from
different populations. In 1992, the Dombrock blood group
system was expanded to include three phenotypically
related antigens, Gya, Hy, and Joa, when it was discovered
that the Gy(a–) phenotype was the null of the Dombrock
system. Based on the knowledge that the Dombrock
glycoprotein is attached to the RBC membrane via a
glycosylphosphatidylinositol linkage and subsequent to
the assignment of the corresponding gene to the short
arm of chromosome 12, expressed sequence tags from
terminally differentiating human erythroid cells were
analyzed in silico to identify the DO gene. This allowed
determination of the molecular basis of the various Do
phenotypes and the realization that DO
is identical to the gene encoding a mono-ADPribosyltransferase, ART4. No enzymatic activity in RBCs
has been demonstrated and the function of this
glycoprotein, on the outside surface of RBCs, has yet to
be determined. This review is a synthesis of our current
knowledge of the Dombrock blood group system.
ABBREVIATIONS: GPI = glycosylphosphatidylinositol; NBF =
National Blood Foundation; SNP(s) = single-nucleotide
polymorphism(s).
From the New York Blood Center, New York, New York.
Address reprint requests to: Marion E. Reid, PhD, Director,
Immunohematology Laboratory, New York Blood Center,
310 East 67th Street, New York, NY 10021; e-mail:
[email protected].
This work was funded in part by a NIH Specialized Center of
Research (SCOR) grant in transfusion medicine and biology
(HL54459).
doi: 10.1111/j.1537-2995.2005.00527.x
TRANSFUSION 2005;45:92S-99S.
92S
TRANSFUSION
Volume 45, August 2005 Supplement
T
he application for a National Blood Foundation
(NBF) grant was my first experience in writing a
grant in the National Institute of Health (NIH)
format. Going through the process of thinking,
writing, submitting, receiving, and reporting a NBF grant
(NBF 93-11) provided me with the basis, confidence, and
some preliminary data with which to apply for NIH funding. Through the NIH-Specialized Center of Research
(SCOR) grant mechanism, my research has been funded
since the beginning of 1995. The focus of my research has
been, and still is, to understand blood group antigens and
their interactions with the corresponding antibody to
improve antibody identification and provision of antigennegative RBC components to patients with alloantibodies
to blood group antigens.
One blood group system that has been of special
interest throughout my career, which also illustrates the
value of different techniques for the advancement of
knowledge, is the Dombrock system. Numerous colleagues have contributed to revealing the complexity of
Dombrock. This review will summarize the understanding
that has evolved as techniques were applied and includes
classical hemagglutination (which showed certain characteristics of the antigens and antibodies, a relationship of
Hy to Gya, and the fact that the Gy(a–) phenotype was the
null of the Dombrock blood group system); immunoblotting (which provided a tool to allow further characterization of the Dombrock glycoprotein, including the fact that
it is linked to the RBC membrane by a glycosylphosphatidylinositol [GPI] anchor); in silico analysis (which led to
cloning of the gene); manual polymerase chain reaction
(PCR)-based assay (which identified the single-nucleotide
polymorphisms [SNPs] associated with the five antigens
and provided a means to screen for antigen-negative
donors); transfection and hybridoma technology (for
production of monoclonal antibodies [MoAbs]); and
microarray technology (which provides a means to perform high-throughput testing of donor blood and the
identification of two new alleles).
CHARACTERISTICS OF DOMBROCK AS
REVEALED BY HEMAGGLUTINATION
Some 20 years after the indirect antiglobulin test (IAT) was
introduced for pretransfusion testing,1 anti-Doa was iden-
DOMBROCK BLOOD GROUP
TABLE 1. The Dombrock blood group system
RBC Phenotype
Do(a+b–)
Do(a+b+)
Do(a–b+)
Gy(a–)
Hy–
Jo(a–)
Doa
+
+
0
0
0
Weak
Reactivity with anti
Dob
Gya
Hy
0
+
+
+
+
+
+
+
+
0
0
0
Weak
Weak
0
0/weak
+
Weak
Joa
+
+
+
0
0/weak
0
White
persons
18
49
33
Rare
Not found
Not found
Black
persons
11
44
45
163
Rare
Rare
Japanese
persons
1.5
22
76.5
Rare
Not found
Not found
Thai
persons
0.5
13
86.5
Not found
Not found
Not found
Chinese
persons
One (unpublished)
Not found
Not found
TABLE 2. ISBT terminology for the Dombrock blood group system and GenBank accession numbers
Antigen
Doa
Dob
Gya
Hy
Joa
ISBT Name
DO1
DO2
DO3
DO4
DO5
ISBT number
014001
014002
014003
014004
014005
Previous ISBT numbers
206001; 900005
206002; 900011
901004; 900010
GenBank accession number
XM_017877
AF290204; X95826 AF382211-AF382216
AY029514, AY029515, AF461390-92, AF459415-17
HY1 AF382217 to AF382219 HY2 AF382220 to AF382222
AF382223 to AF382225
because RBCs with either the Gy(a–)
phenotype or the Hy– phenotype are
• Resistant to papain or ficin treatment of antigen-positive RBCs
also Jo(a–).13,14 Another high-incidence
• Sensitive to trypsin or DTT (200 mmol/L) treatment of antigen-positive RBCs
antigen, Jca, was shown to be associated
• Weakened by a-chymotrypsin or 2-Aminoethyliso-thiouronium bromide (AET) treatment of
with Gya and Hy15 but later shown not to
antigen-positive RBCs
• Expressed on cord RBCs
be a discrete antigen.16,17
• Absent from paroxysmal nocturnal hemoglobinuria Type III RBCs
It was not until 199218 that a
• Some variation in expression on different RBCs and on RBCs from different people
a
remarkable discovery was made. In
• Not highly immunogenic, with the exception of Gy
• Carried on the Do glycoprotein, a mono-ADP-ribosyltransferase (synonym: ART-4),
addition to being Hy– and Jo(a–), Gy(a–)
which is attached to the RBC membrane by a GPI anchor
RBCs were shown to be Do(a–b–).19
Thus, RBCs with the Gy(a–) phenotype
tified.2 The antibody that recognizes the antithetical antiare the null phenotype of the Dombrock blood group sysgen, anti-Dob, was described 8 years later.3 These two
tem. Upon this discovery, Gya, Hy, and Joa antigens were
antibodies define three phenotypes, the prevalence of
assigned ISBT numbers in the Dombrock blood group
which differs in various ethnicities (see Table 1). For popsystem20,21 (Table 2).
3,4
ulations other than white persons, studies have been
restricted to testing with anti-Doa and, thus, the numbers
SEROLOGIC CHARACTERIZATION OF
given for Do(a–b+) in Table 1 are calculated from the prevDOMBROCK ANTIGENS
a
4-6
a
b
alence given for the Do antigen. Do and Do were
placed in the Dombrock blood group system (DO; 014)
The characteristics of Do antigens are summarized in
by the International Society of Blood Transfusion (ISBT)
Table 3. The susceptibility and resistance of antigens in
Working Party on Terminology for Red Cell Surface Antithe Dombrock system to various proteolytic enzymes,
gens in 1985.7
their sensitivity to treatment of RBCs with dithiothreitol
The high-prevalence antigens Gregory (Gya) and Hol(DTT), and their absence from paroxysmal nocturnal
ley (Hy), which were described independently in 1967,8,9
hemoglobinuria Type III RBCs22 can be used to aid the
10
were shown 8 years later to be phenotypically related.
identification of corresponding antibodies.
Red blood cells (RBCs) from Caucasian persons with the
Gy(a–) phenotype are Hy–, and RBCs from black people of
SEROLOGIC CHARACTERIZATION AND
African descent with the Hy– phenotype are Gy(a+w).10
CLINICAL SIGNIFICANCE OF DOMBROCK
a
Based on this observation, Gy and Hy were upgraded
ANTIBODIES
from the ISBT Series of High Incidence Antigens and
11
placed in the Gregory Collection.
Characteristics of antibodies to antigens in the Dombrock
The high-prevalence antigen Joa12 was shown to have
blood group system, which are summarized in Table 4,
a phenotypic association with Gya and Hy antigens
explain why they can be difficult to identify. The paucity
TABLE 3. Characteristics of antigens in the Dombrock blood group system
Volume 45, August 2005 Supplement
TRANSFUSION
93S
REID
TABLE 4. Characteristics of antibodies to Dombrock antigens
• Usually IgG
• React optimally by column agglutination technology or by the IAT with papain- or
ficin-treated RBCs
• Usually weakly reactive
• Do not bind complement
• Stimulated by pregnancy and by transfusion
• Usually present in sera containing other alloantibodies, with the exception of anti-Gya,
which can occur as a single specificity
• Often deteriorate in vitro and fall below detectable levels in vivo
• Have not caused clinical HDN (positive DAT only) but have caused transfusion reactions
of reliable monospecific antisera hampered studies
involving the Dombrock blood group system.
Although transfusion reactions attributed to antiDoa or anti-Dob have been reported,23-32 they may be
underreported. One reason is that events usually associated with transfusion reactions may not be observed.
For example, the direct antiglobulin test (DAT) is often
negative, antibody may not be eluted from the patient’s
RBC samples after transfusion, the lag phase may be
absent, and there may be no increase in titer of the
antibody.
At least one anti-Hy has caused biphasic destruction
of Hy+ RBCs;33 other examples of anti-Hy (and also antiGya and anti-Joa) have caused moderate transfusion reactions. Antibodies in the Dombrock blood group system
have not been reported to cause clinical hemolytic disease
of the newborn (HDN) although RBCs of some antigenpositive babies were positive in the DAT.
PROPERTIES OF THE DOMBROCK
GLYCOPROTEIN REVEALED BY
IMMUNOBLOTTING
Immunoblotting showed that Gya, Hy, and Joa antigens are
located on the same glycoprotein.34,35 Joa was assigned a
number in the ISBT Series of High Incidence antigens
before its association with Gya and Hy was realized.7 The
antigen bypassed joining the Gregory collection to be
assigned a number in the Dombrock system. When it was
recognized that the Gy(a–) phenotype was the null of
the Dombrock system,19 the Gregory glycoprotein was
renamed the Dombrock (Do) glycoprotein.
The Do glycoprotein has an apparent Mr of approximately 47,000 to 58,000 in sodium dodecyl sulfate–polyacrylamide gel electrophoresis under nonreducing
conditions19,34 and is attached to the RBC membrane by a
GPI linkage.22,34,35 It has several potential N-linked glycosylation sites, and four or five cysteine residues in the
membrane-bound protein.36 The susceptibility of Do antigens to sulfydryl compounds suggests that the tertiary
conformation of the glycoprotein is dependent on disulfide bonds.
94S
TRANSFUSION
Volume 45, August 2005 Supplement
IN SILICO ANALYSIS AND
CLONING THE GENE
ENCODING THE DOMBROCK
GLYCOPROTEIN
Based on the knowledge that the Do glycoprotein is attached to the RBC membrane via a GPI linkage, and on the
assignment of the DO to the short
arm of chromosome 12,37 expressed
sequence tags from terminally differentiating human erythroid cells were analyzed in silico. A candidate gene, DOK1, was detected
within a BAC clone (GenBank Accession Number
AC007655) and, by a series of experiments, was proven to
be the gene that encodes the Do glycoprotein36 (GenBank
Accession Number AF290204). The DOK1 gene is identical
to the ART4 gene (GenBank Accession Number X95826)38
and has been renamed DO (GenBank Accession Number
XM_017877). The Do glycoprotein (ART4) is a member
of the mono-ADP-ribosyltransferase family, but has no
demonstrable enzyme activity on the RBC.39,40
The DO gene consists of three exons distributed over
14 kb of DNA. The messenger RNA is predicted to encode
a protein of 314 amino acids that has both a signal
sequence and a GPI-anchor motif.36 Computer prediction
programs suggested that Do glycoprotein has a signal peptide of 44 amino acids,36 which would be unusually long.
The possibility exists that initiation occurs at the second
AUG and that Met22 to Thr44 is the signal peptide (G.
Cross, personal communication). Indeed, Met1 is not conserved in the mouse, whereas Met22 is conserved.38,41 In
either case, the N-terminal amino acid in the membranebound protein is predicted to be Glu45. The predicted GPIanchor motif of 17 amino acids (residues 298-314)36 is less
than the minimum length found in other GPI-anchored
proteins in mammalian cells,42-45 and it is possible that the
C-terminal 30 amino acids (residues 285-314) forms the
GPI-anchor motif (G. Cross, personal communication).
Thus, the membrane-bound Dombrock glycoprotein is
likely to consist of 240 amino acids (residues 45-284)
rather than 253 amino acids (residues 45-297) (Fig. 1).
In this review, the numbering used by Gubin and
coworkers36 is used, that is, nucleotide 1 is the A of the first
AUG initiation codon and amino acid 1 is the first
methionine.
MANUAL PCR-BASED ASSAYS REVEAL
SNPs AND PROVIDE A MEANS TO FIND
ANTIGEN-NEGATIVE BLOOD
The antithetical antigens, Doa and Dob
The DOA and DOB alleles differ at three nucleotide positions. Two are silent mutations (378C>T; 126Tyr and
624T>C; 208Leu) and one is a missense mutation (793A>G;
DOMBROCK BLOOD GROUP
a slightly higher prevalence than
reported. It has been observed by this
author that provision of Do(b–) blood
as determined by DNA analysis has
improved RBC survival in patients with
anti-Dob.
The high-prevalence Hy antigen
Fig. 1. Diagram showing the molecular basis of Dombrock alleles and the expected Do a
and Dob types.
The nucleotide change associated with
Hy+/Hy– is 323G>T, which is predicted
to encode Gly108Val. The change is
associated with the absence of the Hy
antigen and is on an allele carrying 378C
(DOA), 624C (DOB), and 793G (DOB). Its
association with 793G (265Asp) explains
why RBCs with the Hy– phenotype are
invariably Do(a–b+) (Fig. 2). There are
two forms of the allele encoding the Hy–
phenotype, one with 898G (named HY1
because it was found in the sister of the
original Hy– proband) encoding 300Val
and the other with 898C (named HY2)
encoding 300Leu (which is present on
Hy+ wild type;17 Table 5).
TABLE 5. Differences between human DO alleles
Allele
DOA
DOA-HA
DOB
DOB-SH
HY1
HY2
JO
GY5
nt (aa)*
323 (108)
G (Gly)
G (Gly)
G (Gly)
G (gly)
T (Val)
T (Val)
G (Gly)
G (Gly)
nt (aa)
350 (117)
C (Thr)
C (Thr)
C (Thr)
C (Thr)
C (Thr)
C (Thr)
T (Ile)
C (Thr)
nt
378
C
T
T
C
C
C
T
T
nt
624
T
T
C
C
C
C
T
T
nt (aa)
793 (265)
A (Asn)
A (Asn)
G (Asp)
G (Asp)
G (Asp)
G (Asp)
A (Asn)
A (Asn)
* nt = nucleotide; aa = amino acid.
Asn265Asp), which encodes, respectively, Doa and Dob36
(Table 5 and Fig. 2). These three mutations each can be
readily differentiated by PCR-restriction fragment length
polymorphisms (RFLP), with DraIII for 378C>T, MnlI for
624T>C,46 and BSe RI47 for 793A>G, or by allele-specific
PCR.48
The ability to distinguish DOA from DOB makes it
feasible to type patients and blood donors. This is a tremendous advantage, because owing to the paucity of reliable reagents, screening for large numbers of Do(a–) or
Do(b–) blood donors by hemagglutination has not been
possible. PCR-RFLP analysis has shown that previously
typed Do(a+b–) donors have both DOA and DOB alleles.
Although RBCs from some of these donors have been
shown to have Do(a+b+W) RBCs, others do not express
Dob. Based on these results, the Dob antigen may have
nt (aa)
898 (300)
C (Leu)
C (Leu)
C (Leu)
C (Leu)
G (Val)
C (Leu)
C (Leu)
C (Leu)
The high-prevalence Joa antigen
A single-nucleotide change of 350C>T is
predicted to encode isoleucine at amino
acid residue 117. Nucleotide 350T is
associated with the absence of the Joa
antigen and is on an allele carrying 378T
(DOB), 624T (DOA), and 793A (DOA)
(Table 5; Fig. 2). The genotype of people
whose RBCs have the Jo(a–) phenotype
can be JO/JO or HY/JO.17
Gy(a–) or Donull phenotype
To date, four molecular bases causing silencing of the DO
have been described. Two of them, a mutation in the
donor splice site49 and a mutation in the acceptor splice
site,50 lead to outsplicing of exon 2. A third proband has a
deletion of eight nucleotides within exon 2 that leads to a
frameshift and a premature stop codon51 and the fourth
mechanism is a nonsense mutation in a novel allele [350C;
378T (DOB); 624T (DOA); 793A (DOA)] that has been
named GY549 (Table 6). There is no known pathology associated with the Doa form of the Dombrock glycoprotein,
which has RGN in place of RGD (a motif within adhesion
ligands that is commonly involved in cell-to-cell interactions involving integrin binding). Furthermore, no pathology has been observed with an absence of the entire Do
glycoprotein [Gy(a–)].
Volume 45, August 2005 Supplement
TRANSFUSION
95S
REID
TRANSFECTION AND HYBRIDOMA
TECHNOLOGY
The ability to transfect cells with DO cDNA provided a tool
to immunize mice as the first step to production of MoAbs
to Dombrock antigens.52,53 Although MoAb anti-Dob is
useful to screen for Do(b–) blood donors, it has the same
limitations as human anti-Dob, and it is recommended to
confirm the type by DNA-based analyses. MoAb anti-Do
recognizes epitopes on RBCs from humans and other apes
but not from monkeys, rabbits, dogs, sheep, or mice.53
These MoAbs anti-Do have also made it possible for the
Do glycoprotein to be assigned the cluster of differentiation number CD297.54
MICROARRAY TECHNOLOGY
The ability to detect SNPs on microarrays or bead arrays
provides a means to revolutionize the way we provide
antigen-negative blood in transfusion medicine. This
approach not only provides high-throughput analysis of
multiple SNPs on one chip but also electronically downloads the massive amount of data generated directly to a
database.55-58 This eliminates two labor-intensive steps
that are currently being used for the identification of antigen-negative blood donors. In testing DNA samples from
distinct ethnic groups (69 Ashkenazi Jew, 58 Chinese, 188
Israeli, 100 Thai, and 218 predominantly black African
American New York donors) with a BeadChip array that
included probes for five DO SNPs, two new DO alleles
were found. One, DOB-SH, is the same as HY at
nucleotides 378, 624, and 793 but has wild-type nucleotide 323 and is predicted to express Dob antigen (the
appropriate alleles, i.e., DOA/DOB-SH or DOB-SH/DOBSH, have not yet been found and, thus, testing has not
been possible). The other, DOA-HA, is the same as JO at
nucleotides 378, 624, and 793 but has wild-type nucleotide 350 (Table 5; Fig. 1) and has been shown to express
Doa antigen.59 DOA-HA was found in all five groups,
whereas DOB-SH was only found in African American persons. Even though the number of samples in each ethnic
group was relatively small, the prevalences of the alleles
found are given in Table 7 (the numbers for the New York
donors is not given because they were selected and not
random).59,60
PERSPECTIVES
Fig. 2. Diagrammatic figure showing amino acids associated
with the Doa/Dob, Hy+/Hy–, and Jo(a +)/Jo(a–) polymorphisms
and the relative positions of the antigens on the Do
glycoprotein.
The molecular basis associated with Hy– and Jo(a–) phenotypes was determined only after numerous samples
were analyzed. During this analysis, it became clear that
RBC samples had been misidentified. The most significant
was that of SJ after whom the Joseph phenotype and Joa
antigen were named. Surprisingly, DNA from SJ is
homozygous for the HY allele (HY1/HY2).17 Thus, any
“anti-Joa” identified by use of SJ RBCs is more likely to be
anti-Hy. The subsequent use of such RBCs and antibodies
led to the inadvertent incorrect labeling of reagents. Some
samples have two different rare alleles, which also leads
to confusing serologic test results. For instance, whereas
RBCs from people with the JO/JO or the JO/HY genotype
TABLE 6. Molecular basis for Gy(a–) phenotype
Molecular basis
a>g in an acceptor splice site in intron 2, leading to skipping of exon 2
t>c in donor splice site in intron 2, leading to skipping of exon 2
Deletion of nucleotides 343-350 in exon 2, frameshift and premature stop codon
442C>T in exon 2, Gln148Stop
96S
TRANSFUSION
Volume 45, August 2005 Supplement
Associated allele
DOB
DOB
DOA
GY5
Proband
GY1, GY2, GY350
GY449
HJ51
GY549
DOMBROCK BLOOD GROUP
are Jo(a–), the latter would be expected to have a slightly
weaker expression of the Hy antigen and to be Do(b+),
albeit weakly. The various combinations of alleles and
expected corresponding antigen expression are given in
Table 8. It is becoming apparent that further complexities
exist because people with, for example, HY/DOB have
made anti-Hy and those with JO/DOB have made anti-Joa.
To correctly identify anti-Hy and anti-Joa, as well as other
Dombrock antibodies, it is of value to use reagent RBCs
that have been typed by DNA analysis.
It is apparent that the immune response in different
people with the same Dombrock phenotype varies. One
explanation is that Hy– patients can produce anti-Doa in
addition to anti-Hy, and Jo(a–) patients can produce antiDob in addition to anti-Joa. In contrast, a patient with a HY/
JO genotype, who would be expected to have RBCs that
type Do(a+wb+w) Gy(a+) Hy(a+w) Jo(a–) and would only be
able to make anti-Joa. This helps to explain some of the
observed variations in reaction strength when testing
plasma and/or serum from people with the same appar-
TABLE 7. Dombrock (DO) alleles in ethnic populations
Alleles
DOA/DOA
DOA/DOA-HA
DOA/HY
DOA/DOB
DOB/DOA-HA
DOB/DOB
Ashkenazi
(% in 69)
13.1
1.4
0
30.4
2.9
52.2
Chinese
(% in 58)
0
1.7
0
12.1
6.9
79.3
Israeli
(% in 188)
11.2
5.3
0.5
42.6
5.3
35.1
Thai
(% in 100)
2
0
0
14
11
73
ent phenotype (see Table 7). The results of DNA analysis
of unusual Do phenotypes, as defined by hemagglutination, has provided an explanation for the diversity of RBC
typing results in antibody producers and for the diversity
of the reactivity of antibodies in their plasma and/or
serum.
The close proximity of amino acids associated with
Hy (residue 108) and Joa (residue 117) antigen expression
may explain why Hy– RBCs appear to lack the Joa antigen
(or express it extremely weakly61) and why Jo(a–) RBCs
have a weak expression of Hy (Fig. 2). The two critical residues are separated by only eight amino acids, which is
within the range of an antigenic determinant.62 In contrast, given the distance of Hy and Joa from, respectively,
Dob or Doa, the reason for weak expression of Dob on RBCs
with the Hy– phenotype and the weak expression of Doa
on RBCs with the Jo(a–) phenotype is still not understood
but presumably is due to conformation. The reason for the
weak expression of Gya antigen on Hy– RBCs is also not
understood.
Determination of the molecular basis underlying the
antigens in the Dombrock blood group system has several
advantages and can potentially change practice in transfusion medicine. The first is the possibility of use of RBCs
with a bona fide antigen profile in antibody identification
panels. The second advantage is to type patients to aid in
antibody identification. Third is the ability to type donors
for DOA and DOB and thereby, perhaps for the first time,
to reliably select Do(b–) RBCs for transfusion to a patient
who has or has had anti-Dob. Because antibodies to antigens in the Do blood group system (especially anti-Dob)
TABLE 8. Expected expression of Do antigens encoded by various combinations of alleles
Phenotype
Common phenotypes
Gy(a+w) Hy– Jo(a–)
Gy(a+) Hy+w Jo(a–)
Other
Alleles present
DOA/DOA
DOA/DOB
DOB/DOB
HY/HY†
HY/JO
JO/JO
DOA/HY
DOB/HY
DOA/JO
DOB/JO
DOA/DOA-HA
DOA-HA/DOA-HA
DOA-HA/DOB
DOA-HA/HY
DOA-HA/JO
DOA/DOB-SH
DOB/DOB-SH
DOB-SH/DOB-SH
DOB-SH/HY
DOB-SH/JO
Doa
2+
1+
0
0
±
1+
1+
0
1+
±
2+
2+
1+
1+
2+
1+
0
0
0
±
RBC known or predicted
Dob
Gya
Hy
0
2+
2+
1+
2+
2+
2+
2+
2+
1+
±
0
±
1+
w
0
2+
±
±
1+
1+
1+
1+
1+
0
2+
1+S
±
2+
1+S
0
2+
2+
0
2+
2+
1+
2+
2+
±
2+
2+
0
2+
2+
1+
2+
2+
2+
2+
2+
2+
2+
2+
1+
2+
2+
1+
2+
2+
Joa
2+
2+
2+
0/w
0
0
1+
1+
1+
1+
2+
2+
2+
2+
1+
2+
2+
2+
2+
2+
Possible antibodies
to Dombrock antigens*
Anti-Dob
Anti-Doa
Anti-Hy; anti-Joa; anti-Doa
Anti-Joa
Anti-Joa; anti-Dob
(Anti-Hy)
(Anti-Hy)
Anti-Dob
Anti-Dob
Anti-Dob
Anti-Dob
Anti-Doa
Anti-Doa
Anti-Doa
* Antibodies in parentheses, even though unexpected, have been described, indicating other reasons (yet to be identified) for silencing Do
alleles exist.
† Because HY1 and HY2 encode the same phenotype, HY is used.
Volume 45, August 2005 Supplement
TRANSFUSION
97S
REID
are rarely available as a single specificity with strength and
volume to make accurate typing possible, this is, perhaps,
the first instance in which DNA-based analyses are more
reliable than hemagglutination.
ACKNOWLEDGMENTS
I am indebted to the numerous colleagues who have allowed me
to be included in their studies that made it possible for me to be
involved in unraveling the complexities of the Dombrock blood
group system. They made it possible for me to be a part of the
investigations with different technologies for more than 30 years
(1974-2005). I am also grateful to the NBF for providing funding
that helped me along the road to grant-funded research. I thank
Christine Lomas-Francis for helpful comments and Robert Ratner for assistance in preparation of the manuscript and figures.
REFERENCES
1. Coombs RR, Mourant AE, Race RR. A new test for detection
of weak and “incomplete” agglutinins. Br J Exp Pathol
1945;26:255-9.
2. Swanson J, Polesky HF, Tippett P, Sanger R. A “new” blood
group antigen, Doa [abstract]. Nature 1965;206:313.
3. Molthan L, Crawford MN, Tippett P. Enlargement of the
Dombrock blood group system. the finding of anti-Dob. Vox
Sang 1973;24:382-4.
4. Polesky HF, Swanson JL. Studies on the distribution of
the blood group antigen Doa (Dombrock) and the
characteristics of anti-Doa. Transfusion 1966;6:294-6.
5. Nakajima H, Moulds JJ. Doa (Dombrock) blood group
antigen in the Japanese: tests on further population and
family samples. Vox Sang 1980;38:294-6.
6. Chandanayingyong D, Sasaki TT, Greenwalt TJ. Blood
groups of the Thais. Transfusion 1967;7:269-76.
7. Lewis M, Allen FH Jr, Anstee DJ, et al. ISBT Working Party on
Terminology for Red Cell Surface Antigens: Munich report.
Vox Sang 1985;49:171-5.
8. Swanson J, Zweber M, Polesky HF. A new public antigenic
determinant Gya (Gregory). Transfusion 1967;7:304-6.
9. Schmidt RP, Frank S, Baugh M. New antibodies to high
incidence antigenic determinants (anti-So, anti-E1, anti-Hy
and anti-Dp) [abstract]. Transfusion 1967;7:386.
10. Moulds JJ, Polesky HF, Reid M, Ellisor SS. Observations on
the Gya and Hy antigens and the antibodies that define
them. Transfusion 1975;15:270-4.
11. Lewis M, Anstee DJ, Bird GW, et al. Blood group terminology
1990: ISBT Working Party on terminology for red cell surface
antigens. Vox Sang 1990;58:152-69.
12. Jensen L, Scott EP, Marsh WL, et al. Anti-Joa: an antibody
defining a high-frequency erythrocyte antigen. Transfusion
1972;12:322-4.
13. Weaver T, Kavitsky D, Carty L, et al. An association between
the Joa and Hy phenotypes [abstract]. Transfusion 1984;24:
426.
98S
TRANSFUSION
Volume 45, August 2005 Supplement
14. Brown D. Reactivity of anti-Joa with Hy– red cells [abstract].
Transfusion 1985;25:462.
15. Laird-Fryer B, Moulds MK, Moulds JJ, Johnson MH, Mallory
DM. Subdivision of the Gya-Hy phenotypes [abstract].
Transfusion 1981;21:633.
16. Weaver T, Lacey P, Carty L. Evidence that Joa and Jca are
synonymous [abstract]. Transfusion 1986;26:561.
17. Rios M, Hue-Roye K, Øyen R, Miller J, Reid ME. Insights into
the Holley-negative and Joseph-negative phenotypes.
Transfusion 2002;42:52-8.
18. Banks JA, Parker N, Poole J. Evidence to show that Dombrock
(Do) antigens reside on the Gya/Hy glycoprotein [abstract].
Transfus Med 1992;2(Suppl 1):68.
19. Banks JA, Hemming N, Poole J. Evidence that the Gya, Hy and
Joa antigens belong to the Dombrock blood group system.
Vox Sang 1995;68:177-82.
20. Daniels GL, Moulds JJ, Anstee DJ, et al. ISBT Working Party
on Terminology for Red Cell Surface Antigens: Sao Paulo
report. Vox Sang 1993;65:77-80.
21. Daniels GL, Anstee DJ, Cartron JP, et al. Blood group
terminology 1995: ISBT Working Party on terminology for
red cell surface antigens. Vox Sang 1995;69:265-79.
22. Telen MJ, Rosse WF, Parker CJ, Moulds MK, Moulds JJ.
Evidence that several high-frequency human blood group
antigens reside on phosphatidylinositol-linked erythrocyte
membrane proteins. Blood 1990;75:1404-7.
23. Kruskall MS, Greene MJ, Strycharz DM, Getman EM, Cawley
A. Acute hemolytic transfusion reaction due to antiDombrocka (Doa) [abstract]. Transfusion 1986;26:545.
24. Polesky HF, Swanson J, Smith R. Anti-Doa stimulated by
pregnancy. Vox Sang 1968;14:465-6.
25. Webb AJ, Lockyer JW, Tovey GH. The second example of
anti-Doa. Vox Sang 1966;11:637-9.
26. Williams CH, Crawford MN. The third examples of anti-Doa.
Transfusion 1966;6:310.
27. Moulds J, Futrell E, Fortez P, McDonald C. Anti-Doa—further
clinical and serological observations [abstract]. Transfusion
1978;18:375.
28. Judd WJ, Steiner EA. Multiple hemolytic transfusion
reactions caused by anti-Doa. Transfusion 1991;31:477-8.
29. Roxby DJ, Paris JM, Stern DA, Young SG. Pure anti-Doa
stimulated by pregnancy. Vox Sang 1994;66:49-50.
30. Moheng MC, McCarthy P, Pierce SR. Anti-Dob implicated as
the cause of a delayed hemolytic transfusion reaction.
Transfusion 1985;25:44-6.
31. Halverson G, Shanahan E, Santiago I, et al. The first reported
case of anti-Dob causing an acute hemolytic transfusion
reaction. Vox Sang 1994;66:206-9.
32. Strupp A, Cash K, Uehlinger J. Difficulties in identifying
antibodies in the Dombrock blood group system in
multiply alloimmunized patients. Transfusion 1998;
38:1022-5.
33. Beattie KM, Castillo S. A case report of a hemolytic
transfusion reaction caused by anti-Holley. Transfusion
1975;15:476-80.
DOMBROCK BLOOD GROUP
34. Spring FA, Reid ME. Evidence that the human blood
group antigens Gya and Hy are carried on a novel
glycosylphosphatidylinositol-linked erythrocyte membrane
glycoprotein. Vox Sang 1991;60:53-9.
35. Spring FA, Reid ME, Nicholson G. Evidence for expression
of the Joa blood group antigen on the Gya/Hy-active
glycoprotein. Vox Sang 1994;66:72-7.
36. Gubin AN, Njoroge JM, Wojda U, et al. Identification of the
Dombrock blood group glycoprotein as a polymorphic
member of the ADP-ribosyltransferase gene family. Blood
2000;96:2621-7.
37. Eiberg H, Mohr J. Dombrock blood group (DO): assignment
to chromosome 12p. Hum Genet 1996;98:518-21.
38. Koch-Nolte F, Haag F, Braren R, et al. Two novel human
members of an emerging mammalian gene family related to
mono-ADP-ribosylating bacterial toxins. Genomics 1997;
39:370-6.
39. Koch-Nolte F, Haag F. Mono (ADP-ribosyl) transferases and
related enzymes in animal tissues. In: Haag F, Koch-Nolte F,
editors. ADP ribosylation in animal tissues. structure,
function, and biology of mono (ADP-ribosyl) transferases
and related enzymes. New York: Plenum Press; 1997. p. 1-13.
40. Okazaki IJ, Moss J. Characterization of
glycosylphosphatidylinositiol-anchored, secreted, and
intracellular vertebrate mono-ADP-ribosyltransferases.
Annu Rev Nutr 1999;19:485-509.
41. Glowacki G, Braren R, Cetkovic-Cvrlje M, et al. Structure,
chromosomal localization, and expression of the gene for
mouse ecto-mono (ADP-ribosyl) transferase ART5. Gene
2001;275:267-77.
42. Udenfriend S, Kodukula K. Prediction of omega site in
nascent precursor of glycosylphosphatidylinositol protein.
Methods Enzymol 1995;250:571-82.
43. Bucht G, Wikstrom P, Hjalmarsson K. Optimising the signal
peptide for glycosyl phosphatidylinositol modification of
human acetylcholinesterase using mutational analysis and
peptide-quantitative structure-activity relationships.
Biochim Biophys Acta 1999;1431:471-82.
44. Bucht G, Hjalmarsson K. Residues in Torpedo californica
acetylcholinesterase necessary for processing to a glycosyl
phosphatidylinositol-anchored form. Biochim Biophys Acta
1996;1292:223-32.
45. Coyne KE, Crisci A, Lublin DM. Construction of synthetic
signals for glycosyl-phosphatidylinositol anchor
attachment: analysis of amino acid sequence requirements
for anchoring. J Biol Chem 1993;268:6689-93.
46. Rios M, Hue-Roye K, Lee AH, et al. DNA analysis for the
Dombrock polymorphism. Transfusion 2001;41:1143-6.
47. Vege S, Nance S, Westhoff C. Genotyping for the Dombrock
DO1 and DO2 alleles by PCR-RFLP [abstract]. Transfusion
2002;42(Suppl):110S-111S.
48. Wu GG, Jin ZH, Deng ZH, Zhao TM. Polymerase chain
reaction with sequence-specific primers-based genotyping
of the human Dombrock blood group DO1 and DO2 alleles
and the DO gene frequencies in Chinese blood donors. Vox
Sang 2001;81:49-51.
49. Rios M, Storry JR, Hue-Roye K, Chung A, Reid ME. Two new
molecular bases for the Dombrock null phenotype. Br J
Haematol 2002;117:765-7.
50. Rios M, Hue-Roye K, Storry JR, et al. Molecular basis of the
Dombrock null phenotype. Transfusion 2001;41:1405-7.
51. Lucien N, Celton JL, Le Pennec PY, Cartron JP, Bailly P. A
short deletion within the blood group Dombrock locus
causing a Donull phenotype. Blood 2002;100:1063-4.
52. Halverson GR, Schawalder A, Miller JL, Reid ME. Production
of a monoclonal antibody shows the conservation of
epitopes on the Dombrock glycoprotein in the great apes
[abstract]. Transfusion 2001;41(Suppl):24S.
53. Mogos C, Schawalder A, Halverson GR, Reid ME. Studies on
the Dombrock blood group system in non-human primates.
Immunohematology 2003;19:77-82.
54. Parusel I, Kahl S, Braasch F, et al. A panel of monoclonal
antibodies recognizing GPI-anchored ADPribosyltransferase ART4, the carrier of the Dombrock blood
group antigen. Immunol Lett 2005;in press.
55. Maaskant-van Wijk PA, Feskens M, Tomson B, Van Rhenen
DJ. Medium-throughput HPA genotyping enables blood
banks to provide matched platelets on short notice
[abstract]. Vox Sang 2004;87(Suppl 3):82.
56. St Louis M, Montpetit A, Phillips MS, Lemieux R. Highthroughput genotyping of blood donors for minor blood
group and major platelet antigens [abstract]. Vox Sang
2004;87(Suppl 3):83.
57. Reid ME, Vissavajjhala P, Hue-Roye K, et al. Blood group
antigen typing using custom Beadchips™ [abstract]. Vox
Sang 2004;87(Suppl 3):82.
58. Denomme GA, Nunes S, Van Oene M. GenomeLab
SNPStream™ ultra-high throughput multiplex SNP analysis
from RBC and platelet genotypes [abstract]. Vox Sang
2004;87(Suppl 3):15.
59. Hashmi G, Shariff T, Seul M, et al. A flexible array format for
molecular diagnostics: Enabling large-scale, rapid blood
group antigen DNA typing. Transfusion 2005;45:680-8.
60. Reid ME, Vissavajjhala P, Palacajornsuk P, et al. Novel
Dombrock blood group genetic variants in different ethnic
groups identified by high throughput BeadChip DNA
analysis [abstract]. Blood 2004;104:113a.
61. Scofield T, Miller J, Storry JR, Rios M, Reid ME. An example
of anti-Joa suggests that Holley-negative red cells do express
some Joa antigen [abstract]. Transfusion 2001;41(Suppl):
24S.
62. Roitt I, Brostoff J, Male DK. Immunology. 6th ed. London:
Mosby; 2001.
63. Smart EA, Fogg P, Abrahams M. The first case of the
Dombrock-null phenotype reported in South Africa
[abstract[. Vox Sang 2000;78(Suppl 1):P015.
Volume 45, August 2005 Supplement
TRANSFUSION
99S