Colton Chido Rogers

Journal of Blood Group Serology and Education
Volume 26, Number 1, 2010

Immunohematology
25
Ce l e b r a t i ng
Ye a r s
Immunohematology
Volume 26, Number 1, 2010
Contents
Letter to the Readers
G.M. Meny
In Memoriam
Marie Cutbush Crookston (19202009)
D. Mallory
In Memoriam
Eloise Giblett (19212009)
D. Mallory
Announcements
Advertisements
Instructions for Authors
1
2
8
11
22
27
30
39
40
41
42
46
Review
The Knops blood group system: a review
J.M. Moulds
Case Report
Role for serial prenatal anti-Vel quantitative serologic monitoring
with 2-ME serum treatment
W.J. Linz, J.T. Fueger, S. Allen, and S.T. Johnson
Review
Granulocyte serology: current concepts and clinical signifcance
M.E. Clay, R.M. Schuller, and G.J. Bachowski
Review
A review of the Colton blood group system
G.R. Halverson and T. Peyrard
Case Report
Alloimmunization to the D antigen by a patient with weak D type 21
H. McGann and R.E. Wenk
Review
A review of the Chido/Rodgers blood group
R. Mougey
Immunohematology is published quarterly (March, June, September, and December) by the American Red Cross, National
Headquarters, Washington, DC 20006.
Immunohematology is indexed and included in Index Medicus and MEDLINE on the MEDLARS system. The contents are also cited in
the EBASE/Excerpta Medica and Elsevier BIOBASE/Current Awareness in Biological Sciences (CABS) databases.
The subscription price is $40.00 (U.S.) and $50.00 (foreign) per year.
Subscriptions, Change of Address, and Extra Copies:
Immunohematology, P.O. BOX 40325, Philadelphia, PA 19106
Or call (215) 451-4902
Web site: www.redcross.org/immunohematology
Copyright 2010 by The American National Red Cross
ISSN 0894-203X
Editors-in-Chief
Sandra Nance, MS, MT(ASCP)SBB
Philadelphia, Pennsylvania
Connie M. Westhoff, PhD, MT(ASCP)SBB
Senior Medical Editor
Geralyn M. Meny, MD
Associate Medical Editors
David Moolten, MD
Ralph R. Vassallo, MD
Emeritus Editorial Board
Delores Mallory, MT(ASCP)SBB
Supply, North Carolina
Patricia Arndt, MT(ASCP)SBB
Pomona, California
James P. AuBuchon, MD
Seattle, Washington
Martha R. Combs, MT(ASCP)SBB
Durham, North Carolina
Geoffrey Daniels, PhD
Bristol, United Kingdom
Anne F. Eder, MD
Washington, District of Columbia
George Garratty, PhD, FRCPath
Pomona, California
Brenda J. Grossman, MD
St. Louis, Missouri
Christine Lomas-Francis, MSc
New York City, New York
Gary Moroff, PhD
Rockville, Maryland
John J. Moulds, MT(ASCP)SBB
Shreveport, Louisiana
Paul M. Ness, MD
Baltimore, Maryland
Joyce Poole, FIBMS
Bristol, United Kingdom
Mark Popovsky, MD
Braintree, Massachusetts
Marion E. Reid, PhD, FIBMS
S. Gerald Sandler, MD
Washington, District of Columbia
Jill R. Storry, PhD
Lund, Sweden
David F. Stroncek, MD
Bethesda, Maryland
Editorial Board
Managing Editor
Cynthia Flickinger, MT(ASCP)SBB
Technical Editors
Christine Lomas-Francis, MSc
Dawn M. Rumsey, ART (CSMLT)
Glen Allen, Virginia
Editorial Assistant
Patti A. Brenner
Copy Editor
Mary L. Tod
Proofreader
Lucy Oppenheim
Electronic Publisher
Wilson Tang
Production Assistant
Marge Manigly
IMMUNOHEMATOLOGY, Volume 26, Number 1, 2010 1
Letter to the Readers
This issue introduces an exciting change to the cover of Immunohematology. Images depicting a visual interpretation of
a journal article or another timely aspect of blood banking will grace the cover of selected issues. I am pleased to announce
that David Moolten, MD, award-winning poet and associate medical editor of Immunohematology, will provide a descrip-
tion of the cover art from both an artistic and a scientifc perspective. I hope these images remind us of the art as well as the
science of transfusion medicine. Your feedback is welcome.
Geralyn M. Meny, MD
Senior Medical Editor
Immunohematology
On Our Cover
El Acueducto de las Ferreras, an aqueduct also known as Puente del Diablo (Devils Bridge), spans a ravine
approximately 4 km north of the town to Tarragona in the Autonomous Community of Catalonia, Spain. The aqueduct
has a maximum height of 26 m and a length of 249 m. Built by the Roman Empire to supply water to the ancient city
of Tarraco, the aqueduct features a superimposed series of arches, and may represent the frst aqueduct of its kind.
The Colton blood group antigen (discussed in this issue) is of course found on aquaporin-1, a widely expressed water
channel protein.
David Moolten, MD
Communication
G.M. Meny
The Knops blood group system: a review
The Knops blood group system fnds its roots in the
group of antibodies that were previously called high-titer,
low-avidity, or HTLA, antibodies. This term was used to de-
scribe a group of alloantibodies that appeared to have many
of the same characteristics, including very weak reactions
at the anti-human globulin (AHG) phase of testing. Howev-
er, as their specifcities were more clearly defned, each was
placed into a blood group system, and Knops became the
22nd system recognized by the ISBT Committee on Termi-
nology for Red Cell Surface Antigens.
1
Since that time, the
protein bearing the Knops antigens has been identifed, the
gene has been cloned, and all of the known Knops antigens
have been characterized at the molecular level. This has al-
lowed investigators to more precisely study the role of the
Knops blood group system in disease.
History
In 1965, the sera from three unrelated patients (Cope-
land, Stirling, and Wainright) were reported as having an
antibody with similar specifcity.
2
The antibody was named
after the frst two antibody producers, Co and St (Cost),
and given the designation Cs
a
. Ten years later a new anti-
gen, York (Yk
a
) was reported.
3
The York serum was initially
believed to be an example of anti-Cs
a
until the RBCs of the
donor were found to be Cs(a+). This same study found that
12 of 1,246 Caucasian donors were both
Cs(a) and Yk(a), suggesting that these
antigens were at least phenotypically re-
lated. Neither, however, was given blood
group status, and they became part of the
HTLA group of antibodies. Even though
Yk
a
was thought to be associated with Cs
a

at this time, it would later be assigned to
the Knops system.
Another similar antibody being inves-
tigated in the 1960s was anti-Kn
a
. Anti-Kn
a

was described in a transfused Caucasian
woman who had a saline-reactive anti-K
plus an unidentifed antiglobulin-reactive
antibody to a high-frequency antigen.
4

The antithetical Kn
b
was later reported in
the Hall serum, which contained a potent
anti-Kp
b
and was being used to screen
Australian blood donors.
5
The Knops blood group system began to expand when
Molthan and Moulds
6
described a new antigen, McC
a
, which
seemed to be related to Kn
a
. They reported that 53 percent
of McC(a) samples were also Kn(a). Interestingly, a ma-
jority of McC
a
antibody producers were Black whereas most
of those making anti-Kn
a
were Caucasian, thus suggesting
that ethnic differences might exist in their respective gene
frequencies.
The next pair of alleles, Sl
a
and Vil, was reported in
separate abstracts with one author using the term McC
c
for
Sl
a
and McC
d
for Vil.
7,8
The Sl terminology came from the
names of the frst two antibody producers, i.e., Swain and
Langley. It was preferred by these authors because they be-
lieved that this antigen was independent from McCoy. After
identifcation of the protein bearing the Knops antigens, Sl
a

was renamed S11 and Vil became S12.
9
The last high-incidence antigen identifed in the Knops
system was KAM,
10
later renamed KCAM by the ISBT Com-
mittee on Terminology for Red Cell Surface Antigens.
11
The
antibody producer was a Caucasian man who exhibited the
Helgeson phenotype, i.e., serologic Kn null. Like Sl
a
, this
antigen showed a widely diverse frequency in Blacks vs.
Caucasians (Table 1). Although KCAM is a high-frequency
antigen in Caucasians, only 20 percent of African Blacks
were KCAM+.
Review
J.M. Moulds

Table 1. Frequency of the Knops antigens in several ethnic groups
Population Kn
a
Kn
b
McC
a
McC
b
S11 S12 Yk
a
KCAM Refs.
Caucasians 98% 4% 98% 1% 99% <1% 90% 98% 3, 5, 8,
12, 13
West
Africans
100% NT 8992% 4954% 3038% 95% NT 20% 10, 14
African
Americans
99% <1%* 90% 44% 5161% 80%* 98% NT 8, 13,
14
African
Brazilians
98% 2% 93% 42% 70% 86% NT 53% 12
Asian
Brazilians
100% 0% 100% 0% 100% 0% NT 95% 12
*Marilyn Moulds, personal communication.
Note: The publication by Covas et al.
12
incorrectly lists 4870A
(isoleucine) as being KCAM negative.
NT = not tested.
Knops blood group system
Knops Protein
In 1991, two groups identifed complement receptor one
(CR1) as the protein carrying the Kn
a
, McC
a
, and Sl
a
blood
group antigens.
15,16
In addition, Moulds et al.
15
also located
Yk
a
to CR1 and suggested that the Helgeson phenotype was
attributable to low CR1 copy numbers on the erythrocytes
(E-CR1). Several years later, Petty et al.
17
confrmed these
reports using monoclonal antibody-specifc immobiliza-
tion of erythrocyte antigens (MAIEA) analyses and also
confrmed that Cs
a
was not on CR1. Thus, Cs
a
and Cs
b
have
remained a collection as defned by ISBT.
CR1 is a membrane-bound glycoprotein and, with the
exception of platelets, is found on most human periph-
eral blood cells. Depending on the methods used, erythro-
cytes display approximately 300 to 800 CR1 molecules per
cell whereas leukocytes display approximately 10,000 to
30,000 molecules per cell. Because erythrocytes are pres-
ent in the peripheral circulation at concentrations 10
3
-fold
higher than the peripheral blood mononuclear cells (PB-
MCs), they account for greater than 85 percent of CR1 in
the blood. E-CR1 binds immune complexes (ICs) that are
shuttled to the liver or spleen for transfer to and ingestion
by macrophages, leading to their elimination. IC-free eryth-
rocytes return to the circulation, where they may continue
participating in IC clearance. We have noted that individu-
als with high CR1 copy numbers may exhibit a weak false-
positive DAT in the gel technique, most likely as a result of
increased binding of ICs.
Knops Antigen Characteristics
The Knops antigens can vary greatly in strength (Table
2), and weakly reactive cells can be falsely phenotyped as
negative if the cells are stored for some duration or have
low E-CR1 numbers. Chemicals that can disrupt disulfde
bonds, i.e., DTT and AET, can destroy Knops, McCoy,
Swain-Langley, and York antigens because they destroy the
conformational structure of the short consensus repeats
(SCRs). There are no examples of these antibodies that are
enhanced by enzyme treatment of RBCs, and most are still
reactive (sometimes weaker) with either fcin- or papain-
treated cells.
18,19
However, all Knops system antibodies cur-
rently identifed are nonreactive with trypsin-treated RBCs.
It is known that a trypsin cleavage site exists in SCR 28 of
the CR1 protein. Because the blood group antigens identi-
fed to date have been found in SCR 25, they are lost on
trypsin treatment of the cells. This fact can be a useful tool
not only in antibody identifcation but also for absorption to
remove other antibodies from a sample.
Although the variability in antigen strength can be in-
herited, it may also be acquired. Diseases that have high
levels of ICs being processed, e.g., systemic lupus erythe-
matosus or HIV infection, can cause an increased loss of
E-CR1, resulting in weakened Knops antigen strength.
20
In
Knops Antibodies
The Knops antibody characteristics are shown in Table
3. Two characteristics that were often attributed to Knops
antibodies were (1) they were not neutralized with plasma,
saliva, or urine, and (2) they were diffcult to adsorb and
elute. The latter most likely refects the low density of the
CR1 protein on the RBC membrane. However, Race and
Sanger
23
reported that adsorption performed with buffy
coats (WBCs) were able to remove anti-Kn
a
from serum.
This led to the speculation that anti-Kn
a
and related speci-
fcities were WBC antibodies. The ability of antigen-positive
WBCs to adsorb Knops antibodies most likely relates to the
fact that the CR1 copy number on WBCs is in the tens of
thousands compared with a few hundred on RBCs.
Although CR1 has not been found in the saliva, low
levels have been found in both urine
24
and plasma.
25
These
are believed to be the result of proteolytic cleavage of CR1
from leukocytes.
26
Serum CR1 is present only in nano-
gram amounts,
27
and therefore the levels are insuffcient to
neutralize Knops antibodies using routine serologic tech-
niques. Hence, Moulds and Rowe
28
developed an inhibition
technique using recombinant, soluble CR1 (sCR1). Because
their source of sCR1 was genetically positive for Kn
a
, McC
a
,
Sl
a
, and Yk
a
, it would only inhibit these antibodies and not
inhibit anti-Kn
b
or McC
b
. It must be remembered that the
Knops phenotype of the sCR1 will be dependent on the gene
chosen for its production. More recently, these investigators
addition, when RBCs are stored, vesicles are budded from
the membrane, and these vesicles contain CR1.
21
Therefore,
the longer RBCs are stored, the weaker the Knops antigens
become. This may explain why In(Lu) RBCs were initially
reported to have weakened Knops antigens.
22

Table 2. Characteristics of the Knops antigens
Inherited as Mendelian-dominant traits
High-frequency RBC antigens (except Kn
b
and McC
b
)
Developed on cord blood cells but may be weaker
Weaker on stored RBCs
Not destroyed by fcin or papain
Destroyed by trypsin treatment
Not found on platelets
Found in low levels in plasma but not in urine or saliva

Table 3. Characteristics of Knops system antibodies
Titer is dependent on E-CR1 levels
Can demonstrate variable reactions
Not neutralized by pooled serum or other body fuids
Diffcult to adsorb and elute from RBCs
IgG reacting by AHG technique
Do not bind complement
Usually not clinically signifcant
J.M. Moulds
have used mutated CR1 constructs to produce peptides ca-
pable of inhibiting anti-McC
b
and S12 (Vil).
29
Clinical Importance
Because the Knops antibodies are found in the serum
of multiply transfused individuals, these sera often contain
additional alloantibodies directed at (for example) K, E,
and Duffy antigens. Most of the Knops antibodies are not
considered clinically signifcant because they do not cause
overt hemolytic transfusion reactions or hemolytic disease
of the fetus and newborn. Some examples of Knops sys-
tem antibodies have been studied using in vitro tests such
as the monocyte monolayer assay (MMA). Arndt and Gar-
ratty
30
studied three anti-Kn
a
, three anti-Yk
a
, and fve anti-
Kn/McC and found that only one had a macrophage index
of greater than 20, suggesting increased RBC destruction.
Other MMA studies of two anti-Kn
a
, two anti-Sl
a
, one anti-
KCAM, and one anti-Yk/Cs all gave indexes of less than
0, indicating no enhanced RBC destruction (J.J. Moulds,
unpublished data).
The CR1 Gene
The CR1 gene resides on chromosome 1 (1q32) and
comprises 39 exons spread out over approximately 133
kilobase (kb) pairs of DNA.
31
These exons encode SCRs
of approximately 60 amino acids in the functional CR1
protein. Seven SCRs are organized into larger units called
long homologous repeats (LHRs). The most
common size protein product, CR1-1, is made
up of four LHRs (A, B, C, D), a transmembrane
region (TM), and a cytoplasmic tail (CYT), as
shown in Figure 1. The binding sites for C3b
and C4b have been localized to SCRs 8-9 and
15-16 (LHRs B and C) and to SCRs 1-2 (LHR-A),
respectively.
The CR1 gene also exhibits two other poly-
morphisms besides the Knops blood group. A
structural, or size, polymorphism results from
four different genes encoding four different
molecular weight proteins: 190 kDa (CR1*3), 220 kDa
(CR1*1), 250 kDa (CR1*2), and 280 kDa (CR1*4). Owing to
the looping structure imparted by multiple disulfde bonds,
the molecular weight shifts downward by approximately 30
kDa when the proteins are separated on SDS-PAGE using
nonreducing conditions. It is now known that the molecu-
lar weight polymorphism is independent from the blood
group polymorphism.
The third commonly recognized polymorphism is based
on quantitative differences in E-CR1. A HindIII RFLP is
detected by two allelic fragments of 7.4 or 6.9 kb on South-
ern blots. Homozygotes for the 7.4-kb fragment are high
CR1 expressors (H), and homozygotes for the 6.9-kb frag-
ment are low expressors (L) of CR1.
32
Alternatively, a PCR-
RFLP can be used that results in a band of 1.8 kb for H or
1.3 and 0.5 kb for L alleles.
33
Although this RFLP correlates
with RBC expression in Caucasian and Chinese individu-
als,
34
there is no relationship between this polymorphism
and CR1 expression in African Americans
35
or West Afri-
cans.
36
The exact molecular mechanism regulating CR1
RBC expression has been elusive but does not appear to be
part of the promoter or 3 untranslated portions of the CR1
gene.
37
Molecular Basis of Knops Antigens
Using CR1 deletion constructs, Moulds et al.
29
frst lo-
calized the McCoy and S11 (Sl
a
) antigens to LHR-D of CR1.
By direct DNA sequencing they were then able to identify
two separate mutations in SCR 25 that correlated with these
two blood group antigens. The McC
a
/McC
b
polymorphism
is at base pair 4795, where an A encodes proline (McC
a
)
and a G encodes aspartic acid (McC
b
). The S11/S12 muta-
tion is only 11 amino acids away; at base pair 4828, an A
encodes arginine and a G encodes glycine. Accordingly, the
ISBT has now assigned these antigens to the Knops system
with the numbers shown in Table 4.
38
Two other mutations have been identifed in SCR 25,
one of which was found in a Caucasian and is related to
S11. KMW was initially believed to have produced anti-Sl
a
.
However, after gene sequencing it was found that she was
not only homozygous for S11 but was homozygous for an-
other SNP at position 4855 that substituted threonine for
serine at amino acid 1610. After extensive molecular stud-
ies, Moulds et al.
39
proposed a model in which S13 was
Table 4. Molecular basis of the Knops antigens
ISBT
number Antigen Nucleotide Nucleotide Amino acid
KN1 Kn
a
4708G 4681G Va11561
KN2 Kn
b
4708A 4681A Met1561
KN3 McC
a
4795A 4768A Lys1590
KN4 S11 4828A 4801A Arg1601
KN5 Yk
a
Unknown Unknown Unknown
KN6 McC
b
4795G 4768G Glu1590
KN7 S12 4828G 4801G Gly1601
KN8 S13 (provisional) 4828A, 4855T* 4801A, 4828T* Arg1601, Ser1610
KN9 KCAM 4870A 4843A Ile1615
*Note: The publication by Moulds et al.
39
incorrectly lists the 4855 mutation as
A>G. The correct substitution is T>A.
Nucleotide number when nucleotides are counted beginning with the A of the
AUG start codon
LHR-A LHR-B LHR-C LHR-D TM CYT

SCR 17 SCR 814 SCR 1521 SCR 2227
Fig. 1. Schematic of the CR1-1 protein bearing the
Knops antigens in SCR 25.
a conformational epitope needing S11 (1601R) and S14
(1610S) to react. S14 and the antithetical antigen S15 are
proposed epitopes awaiting the identifcation of the appro-
priate antibodies before they can be offcially recognized.
However, the related SNPs can be identifed molecularly.
Limited population studies of Caucasians, Blacks, and
Asians predict that S14 may be a high-frequency antigen in
all populations and that S15 predominantly occurs in Cau-
casians.
12,39
The fnal antigen assigned to the Knops system is KCAM
(initially reported as KAM). Interestingly, the SNP produc-
ing KCAM was already known but had not been associated
with a blood group specifcity until a McCoy-like antibody
was found in a Caucasian blood donor. On DNA sequenc-
ing, it was discovered that he was homozygous for an SNP
at bp 4870 that substituted valine for isoleucine.
10
Hence,
it was assigned the number KN9, and the antigen was re-
named KCAM (KC for the city where the donor was found
and AM for the donors initials).
Disease Associations
In 1997, Rowe et al.
40
identifed CR1 as a ligand for
the rosetting of Plasmodium falciparuminfected RBCs
with uninfected cells. The ability of erythrocytes infected
with P. falciparum to form rosettes is a property shown by
only some parasite isolates, but is of importance because
it has been associated with severe malaria.
41
These authors
showed that CR1 on uninfected erythrocytes was required
for the formation of rosettes by demonstrating that CR1-
defcient erythrocytes (Helgeson phenotype) had reduced
rosetting and soluble recombinant CR1 could inhibit ro-
setting. RBCs having the Sl:1 phenotype showed reduced
binding to the parasite rosetting ligand PfEMP1.
14
Thus, the
authors hypothesized that this polymorphism may have
been selected for in malaria-endemic regions by provid-
ing protection against severe malaria. The hypothesis was
then tested by studies of Africans in Mali and Kenya, where
it was found that the combined Knops haplotype Kn(a+)/
McC(a)/Sl:1 appeared to be protective.
CR1, as well as other complement receptors, has been
identifed as a receptor facilitating cell entry for a variety
of pathogenic organisms. Pathogens using CR1 include
Leishmania major (monocyte-macrophage),
42
Legion-
ella pneumophila (monocyte-macrophage),
43
Leishmania
panamensis,
44
and Mycobacterium tuberculosis (mono-
cyte-macrophage).
45
In an AABB abstract, Noumsi et al.
46

reported that individuals heterozygous for McC
a
and McC
b

were less likely to be infected by M. tuberculosis (odds ra-
tio, 0.42; 95% confdence interval, 0.22 to 0.81; P = 0.007).
These results suggested that McC
b
may have evolved among
African populations in the context of M. tuberculosis pres-
sure, and conferred a survival advantage in its heterozygous
form.
Conclusions
The molecular identifcation of the Knops blood group
polymorphisms holds the promise for a better means of typ-
ing for these antigens, the only exception to this being the
yet unidentifed Yk
a
. Because of the inherited or acquired
changes in RBC expression of CR1, i.e., Knops antigens,
genotyping may become the method of choice for identify-
ing these antigens, as it is for those of the Dombrock sys-
tem. It is interesting to note that all of the known Knops
polymorphisms are in SCR 25 of the CR1 protein. However,
additional SNPs have been found in the CR1 gene, which
suggests that the identifcation of new Knops blood group
antigens is not yet at an end and will surely provide a chal-
lenge to serologists in the future.
References
Daniels GL, Anstee DJ, Cartron JP, et al. Blood group 1.
terminology 1995. Vox Sang 1995;69:26579.
Giles CM, Huth MC, Wilson TE, Lewis HBM, Grove 2.
GEB. Three examples of a new antibody, anti-Cs
a
,
which reacts with 98% of red cell samples. Vox Sang
1965;10:40515.
Molthan L, Giles CM. A new antigen, Yk 3.
a
, and its rela-
tionship to Cs
a
(Cost). Vox Sang 1975;29:14553.
Helgeson M, Swanson J, Polesky HF. Knops-Helgeson 4.
(Kn
a
), a high frequency erythrocyte antigen. Transfu-
sion 1970;10:7378.
Mallan MT, Grimm W, Hindley L, Knighton G, Moulds 5.
MK, Moulds JJ. The Hall serum: detecting Kn
b
, the
antithetical allele to Kn
a
(abstract). Transfusion
1980;20:630.
Molthan L, Moulds J. A new antigen, McC 6.
a
(McCoy),
and its relationship to Kn
a
(Knops). Transfusion
1978;18:5668.
Lacey P, Laird-Fryer B, Block U, Lair J, Guilbeau L. 7.
Moulds JJ. A new high incidence blood group factor,
Sl
a
; and its hypothetical allele. (abstract).Transfusion
1980;20:632.
Molthan L. Expansion of the York, Cost, McCoy, Knops 8.
blood group system: the new McCoy antigens McC
c
and
McC
d
. Med Lab Sci 1983;40:11321.
Daniels GL, Cartron JP, Fletcher A, et al. International 9.
Society of Blood Transfusion Committee on terminol-
ogy for red cell surface antigens: Vancouver report. Vox
Sang 2003;84:2447.
Moulds JM, Pierce S, Peck KB, Tulley ML, Doumbo O, 10.
Moulds JJ. KAM: a new allele in the Knops Blood Group
System (abstract). Transfusion 2005;45(Suppl):27A.
Daniels G, Castilho L, Flegel WA. International Society 11.
of Blood Transfusion Committee on terminology for
red cell surface antigens: Cape Town report. Vox Sang
2007;92:2503.
Covas DT, de Oliveira FS, Rodrigues ES, Abe-Sandes 12.
K, Silva WA Jr, Fontes AM. Knops blood group hap-
lotypes among distinct Brazilian populations. Transfu-
sion 2007;47:14753.
Molthan L. The serology of the York-Cost-McCoy-Knops 13.
red blood cell system. Am J Med Technol 1983;49:49
56.
Moulds JM, Kassambara L, Middleton JJ. Identifca- 14.
tion of complement receptor one (CR1) polymorphisms
in west Africa. Genes Immun 2000;1:3259.
Moulds JM, Nickells MW, Moulds JJ, Brown M, At- 15.
kinson JP. The C3b/C4b receptor is recognized by the
Knops, McCoy, Swain-Langley, and York blood group
sera. J Exp Med 1991;173:115963.
Rao N, Ferguson DJ, Lee SF, Telen MJ. Identifcation 16.
of human erythrocyte blood group antigens on the C3b/
C4b receptor. J Immunol 1991;146:35017.
Petty AC, Green CA, Poole J, Daniels GL. Analysis of 17.
Knops blood group antigens on CR1 (CD35) by the
MAIEA test and by immunoblotting. Transfus Med
1997;7:5562.
Moulds MK. Serological investigation and clinical sig- 18.
nifcance of high-titer, low-avidity (HTLA) antibodies.
Am J Med Technol 1981;10:78995.
Reid ME, Lomas-Francis C. Knops blood group system. 19.
In: The blood group antigen factsbook. 2nd ed. San Di-
ego, CA: Academic Press, 2004:43954.
Holme E, Fyfe A, Zoma A, Veitch J, Hunter J, Whaley 20.
K. Decreased C3b receptors (CR1) on erythrocytes from
patients with systemic lupus erythematosus. Clin Exp
Immunol 1986;63:418.
Pascual M, Lutz HU, Steiger G, Stammler P, Schifferli 21.
JA. Release of vesicles enriched in complement receptor
1 from human erythrocytes. J Immunol 1993;151:397
404.
Daniels GL, Shaw MA, Lomas CG, Leak MR, Tippett P. 22.
The effect of In(Lu) on some high-frequency antigens.
Transfusion 1986;26:1712.
Race RR, Sanger R. Some very frequent antigens. In: 23.
Blood groups in man. 6th ed. Oxford: Blackwell Scien-
tifc Publications, 1975:41030.
Pascual M, Steiger G, Sadallah S, et al. Identifcation 24.
of membrane-bound CR1 (CD35) in human urine: ev-
idence for its release by glomerular podocytes. J Exp
Med 1994;179:88999.
Yoon SH, Fearon DT. Characterization of a soluble form 25.
of the C3b/C4b receptor (CR1) in human plasma. J Im-
munol 1985;134:33328.
Danielsson C, Pascual M, Larsen B, Steiger G, Schifferli 26.
JA. Soluble complement receptor type 1 (CD35) is re-
leased from leukocytes by surface cleavage. Eur J Im-
munol 1994;24:272531.
Pascual M, Duchosal MA, Steiger G, et al. Circulating 27.
soluble CR1 (CD35). J Immunol 1993;151:170211.
Moulds JM, Rowe KE. Neutralization of Knops system 28.
antibodies using soluble complement receptor 1. Trans-
fusion 1996;36:51720.
Moulds JM, Zimmerman PA, Doumbo OK, et al. Molecu- 29.
lar identifcation of Knops blood group polymorphisms
found in long homologous region D of complement re-
ceptor 1. Blood 2001;97:287985.
Arndt PA, Garratty G. A retrospective analysis of the 30.
value of monocyte monolayer assay results for predict-
ing the clinical signifcance of blood group alloantibod-
ies. Transfusion 2004;44:127381.
Vik DP, Wong WW. Structure of the gene for the F al- 31.
lele of complement receptor type 1 and sequence of
the coding region unique to the S allele. J Immunol
1993;151:621424.
Wilson JG, Murphy EE, Wong WW, Klickstein LB, 32.
Weis JH, Fearon DT. Identifcation of a restriction
fragment polymorphism by a CR1 cDNA that correlates
with the number of CR1 on erythrocytes. J Exp Med
1986;164:509.
Cornillet P, Philbert F, Kazatchkine MD, Cohen JHM. 33.
Genomic determination of the CR1 (CD35) density
polymorphism on erythrocytes using polymerase chain
reaction amplifcation and HindIII restriction enzyme
digestion. J Immunol Methods 1991;136:1937.
Moulds JM, Brai M, Cohen JHM, et al. Reference typ- 34.
ing report for complement receptor 1 (CR1). Exp Clin
Immunogenet 1998;15:2914.
Herrera AH, Xiang L, Martin SG, Lewis J, Wilson JG. 35.
Analysis of complement receptor type 1 (CR1)
expression on erythrocytes and of CR1 allelic markers
in Caucasian and African American populations. Clin
Immunol Immunopathol 1998;87:17683.
Rowe JA, Plowe CV, Diallo DA, et al. Erythrocyte com- 36.
plement receptor 1 expression level does not correlate
with HindIII restriction fragment length polymorphism
in ; implications for studies on malaria susceptibility.
Genes Immun 2002;8:497500.
Cockburn IA, Rowe JA. Erythrocyte complement recep- 37.
tor 1 (CR1) expression level is not associated with poly-
morphisms in the promoter or 3 untranslated regions
of the CR1 gene. Int J Immunogenet 2006;33:1720.
Daniels GL, Anstee DJ, Cartron JP, et al. Interna- 38.
tional Society of Blood Transfusion working party on
terminology for red cell surface antigens. Vox Sang
2001;80:1936.
Moulds JM, Zimmerman PA, Doumbo OK, et al. Ex- 39.
pansion of the Knops blood group system and subdivi-
sion of Sl
a
. Transfusion 2002;42:2516.
Rowe JA, Moulds JM, Newbold CI, Miller LH. 40. P. falci-
parum rosetting mediated by a parasite-variant eryth-
rocyte protein and complement-receptor 1. Nature
1997;388:2925.
J.M. Moulds
Rowe JA, Obeiro J, Newbold CI, Marsh K. 41. Plasmodium
falciparum rosetting is associated with malaria severity
in Kenya. Infect Immun 1995;63:23236.
Da Silva RP, Hall BF, Joiner KA, Sacks DL. CR1, the 42.
C3b receptor, mediates binding of infective Leishmania
major metacyclic promastigotes to human macrophag-
es. J Immunol 1989;143:61722.
Payne NR, Horwitz MA. Phagocytosis of 43. legionella
pneumophila is mediated by human monocyte comple-
ment receptors. J Exp Med 1987;166:137789.
Robledo S, Wozencraft A, Valencia AZ, Saravia N. Hu- 44.
man monocyte infection by Leishmania (viannia) pan-
amensis. J Immunol 1994;152:126575.
Schlesinger LS, Bellinger-Kawahara CG, Payne NR, 45.
Horwitz MA. Phagocytosis of Mycobacterium tuber-
culosis is mediated by human monocyte complement
receptors and complement component C3. J Immunol
1990;144:277180.
Noumsi GT, Tounkara A, Diallo D, Moulds JM. Knops 46.
blood group polymorphism and protection from My-
cobacterium tuberculosis (abstract). Transfusion
2006;46(Suppl):19A.
Joann M. Moulds, PhD, MT(ASCP)SBB, Director, Clinical
Immunogenetics, LifeShare Blood Centers, 8910 Linwood
Avenue, Shreveport, LA 71106.
Attention: State Blood Bank Meeting Organizers
If you are planning a state meeting and would like cop-
ies of Immunohematology for distribution, please send
request, 4 months in advance, to immuno@usa.
redcross.org.
Manuscripts
The editorial staff of Immunohematology welcomes manuscripts pertaining to blood group serology and educa-
tion for consideration for publication. We are especially interested in case reports, papers on platelet and white
cell serology, scientifc articles covering original investigations, and papers on new methods for use in the blood
bank. Deadlines for receipt of manuscripts for consideration for the March, June, September, and December
issues are the frst weeks in November, February, May, and August, respectively. For instructions for scientifc
articles, case reports and review articles, see Instructions for Authors in every issue of Immunohematology or on
the Web at www.redcross.org/immunohematology. Include fax and phone numbers and e-mail address with all
articles and correspondence. E-mail all manuscripts to immuno@usa.redcross.org.
For information concerning Immunohematology,
Journal of Blood Group Serology and Education, or
the Immunohematology Methods and Procedures
manual, contact us by e-mail at immuno@usa.
redcross.org
Role for serial prenatal anti-Vel quantitative
serologic monitoring with 2-ME serum
treatment during pregnancy: case report
W.J. Linz, J.T. Fueger, S. Allen, and S.T. Johnson
Anti-Vel is an uncommon antibody to a high-prevalence antigen.
Its clinical signifcance and management in the prenatal setting are
not well characterized. We present a case that demonstrates the
utility of serial prenatal anti-Vel quantitative serologic monitoring
with 2-ME serum treatment during pregnancy. The patient is a
23-year-old Hispanic woman with history of prior pregnancy and
prior transfusion who was discovered to have an antibody to the
high-prevalence Vel antigen in the frst trimester (week 7) of her
second pregnancy. Interval measurements of the serologic antibody
titers were performed during the next 26 weeks. The untreated se-
rum (IgM and IgG) titer increased from a baseline of 4 to 16 during
that interval, while the 2-ME (presumed IgG component) titer re-
mained stable at 4. Responding to ultrasound fndings suspicious
for fetal anemia, the child was delivered without complications at
34 weeks gestation. At birth, the DAT was negative and there was
no evidence of HDN. Placed in the context of other similar reports,
this case demonstrates the importance of separately reporting the
IgG fraction (after either DTT treatment or 2-ME treatment) from
the untreated (IgM and IgG) fraction and the importance of cor-
relating the treated serum titer with potential clinical signifcance.
Immunohematology 2010;26:0810.
Key Words: Vel antigen collection, anti-Vel, prenatal serologic
monitoring, 2-ME-treated serum, pregnancy
V
el is a high-prevalence antigen frst described by
Sussman et al. in 1952 in association with a trans-
fusion reaction.
1
Although several additional cases
quickly confrmed these authors fndings,
2
our knowledge
of the Vel antigen (collection) remains incomplete. The
presence of the Vel-negative phenotype in the general pop-
ulation is estimated to be 1 in 4000 individuals.
3
Vel alloim-
munization can be mediated through IgG or IgM. Anti-Vel
is thought to be associated rarely with hemolytic disease of
the fetus and newborn (HDFN) because many examples of
anti-Vel are mostly IgM and Vel is weakly expressed on the
surface of fetal RBCs. Although the risk may be low, it is rea-
sonable to monitor prenatal serologic titers serially during
pregnancy.
4
The fndings published in a recent case report
by van Gammeren et al.,
5
of severe HDFN from anti-Vel,
support this approach. The van Gammeren case report de-
scribes discovery in a prenatal patient of an anti-Vel, which
at initial presentation was determined to be composed ex-
clusively of IgM. At the end of the pregnancy, an anti-Vel
titer of 64 was demonstrated in untreated serum, whereas
the DTT-treated serum (presumed IgG content) titer rose
to 16. In their case, the infant exhibited severe jaundice and
reticulocytosis. As a corollary to the van Gammeren case re-
port, we present a case of an anti-Vel for which the IgG titer
as determined by 2-ME treatment was stable through preg-
nancy despite a signifcant rise in the titer of the anti-Vel in
the untreated serum. At birth, the child showed no evidence
of HDFN. This case report supports the importance of serial
determination of the prenatal IgG titer to help assess risk
of HDFN.
Case Report
A 23-year-old Hispanic woman, with a history of prior
blood transfusion and one successful pregnancy, presented
initially early in the frst trimester of the index pregnancy.
Her RBCs were typed as group O, D+. Routine serologic
testing demonstrated variable reactivity using gel technolo-
gy. The autocontrol was negative. Further testing using tube
technology with PEG as enhancement also demonstrated
variable weak reactivity, again with no specifcity. No reac-
tivity was demonstrated when using tube technology with
LISS. A repeat serologic evaluation was performed 4 weeks
later to help clarify the nature of this reactivity. The subse-
quent serologic evaluation demonstrated reactivity with all
RBCs tested whether using gel or test tube technology with
LISS or PEG methods. The autocontrol was consistently
negative. As an antibody to a high-prevalence antigen was
suspected, the sample was referred to a large nationally rec-
ognized immunohematology reference laboratory. With use
of rare RBCs lacking various high-prevalence antigens, re-
activity was determined to be that of anti-Vel. The patients
RBCs were shown to lack the Vel antigen. The initial quan-
titative serologic titer was 4 when tested against Vel+ RBCs
using both 2-ME-treated and untreated serum. The Marsh
score ranged from 14 with untreated serum to 4 with 2-ME-
treated serum. The developing babys RBCs were presumed
to be Vel+ as the fathers RBCs were demonstrated to ex-
press the Vel antigen. Subsequent quantitative serologic
monitoring was performed in approximately 4- to 6-week
intervals for the next 20 weeks of pregnancy (Table 1).
At 33 2/7 weeks gestation, Doppler ultrasonography
showed an increase in middle cerebral artery velocity (>1.5
multiples of the median [MoM] for gestational age) suspi-
cious for fetal anemia. This fnding was a signifcant change
from previous Doppler ultrasound studies and correspond-
ed with the change in quantitative serologic titer of the
Case Report
untreated serum from 4 at presentation to 16 by week 26
(Table 1). A multidisciplinary risk-beneft analysis, includ-
ing consideration of potential maternal and fetal need for
transfusion with an extremely limited blood supply (two
available units), the unpredictable timing issues encoun-
tered with induction, and the limited time-availability for
transfusion after thawing the available units, prompted
the decision to move toward delivery in a manner that op-
timized the availability of blood for the newborn, if it was
indeed anemic. The patient underwent cesarean delivery at
34 3/7 weeks gestational age, 2 days after the initial admin-
istration of Celestone to promote fetal pulmonary maturity.
A 3088-g male newborn with Apgar scores of 7 at 1 min-
ute and 9 at 5 minutes was born without complication. The
childs RBCs typed as group A, D+, and the antibody screen
was negative. Initial laboratory data were also remarkable
for a Hb level of 16.6 g/dL (normal range), reticulocyte count
of 3.5% (normal range), and negative DAT. Approximately
1 week after delivery, a mild rise in bilirubin was observed,
reaching a peak of 10 mg/dL. Given that the DAT and the
initial screen were negative, the mild hyperbilirubinemia
was thought not likely to have resulted from transplacen-
tal anti-Vel or a possible anti-A from a group O mother.
Moreover, although laboratory data are a bit limited, there
was no evidence of anemia or hemolysis. Importantly, the
mild hyperbilirubinemia quickly resolved, and the newborn
was released to home and is doing well approximately 10
months after birth.
performed at the immunohematology reference laboratory
used tube method. The patients serum was diluted in 6%
albumin. A 30-minute 37C incubation was performed, fol-
lowed by a saline IAT (Anti-IgG, ImmucorGamma Inc.).
6

Reaction scoring was performed using the system originally
described by Marsh.
7
The patients serum was treated with
2-ME (Fisher Scientifc Co., Hanover Park, IL) and dialyzed
overnight using dialysis membranes (Spectra/Por2, 12,000
to 14,000 dalton molecular weight cut-off [MWCO], Spec-
trum Laboratories, Inc., Rancho Domingtuez, CA) in PBS,
pH 7.3. Dialyzed serum was tested using the methods previ-
ously described.
Discussion
This case adds to the limited published information
regarding the natural history of anti-Vel in the prenatal
setting, and it supports the value of sorting immunoglobu-
lin classes as an aid to proper prenatal management. It is
worth noting that at initial presentation, weak, variable se-
rologic reactivity was demonstrated. Despite use of several
standard methods, specifcity could not be assigned to the
serologic reactivity. In the initial serologic evaluation, reac-
tivity was present using gel technology and tube technology
with PEG as enhancement. Tube testing with LISS showed
no reactivity. Although a LISS tube method is considered
an appropriate antibody detection method in the prenatal
setting,
8
this method would not have detected the antibody.
A repeat study approximately 7 weeks after the initial study
did demonstrate the presence of strong serologic reactivity
using tube testing with LISS and PEG as well as with gel.
The titer of the untreated serum and 2-MEtreated serum
was 4. Presumably, late in the frst trimester of pregnancy
the mother was exposed to fetal RBCs expressing the Vel
antigen because by week 14 she demonstrated a specifc se-
rologic immune response. Interestingly, for the remainder
of the pregnancy, the titer of the anti-Vel in the 2-ME (pre-
sumed IgG component)treated serum remained stable al-
though the titer in the untreated serum (presumed IgM and
IgG) increased by two dilutions. This observation may be
useful to laboratories without the capacity to treat patient
serum with 2-ME or DTT. Namely, in this case anti-Vel titer
in the untreated serum appeared to rise before the anti-Vel
IgG titer in the 2-MEtreated serum.
The rise in antibody titer in the untreated serum com-
bined with the abnormal Doppler velocimetry prompted
the preterm delivery. At birth, the DAT and antibody screen
were negative, and the hemoglobin and reticulocyte count
were in the normal range. As the anti-Vel present in the ma-
ternal serum contained an IgG component, the negative DAT
suggests that Vel antigen on the fetal RBCs may either be
weakly or partially expressed
9
or the antibody avidity to the
antigen was poor. Regardless, in this case, an anti-Vel titer of
4 does not appear to be independently suffcient for HDFN,
whereas a titer of 16 in the case reported by van Gammeren
et al.
5
was associated with HDFN. The cause of the rise in
Methods
ABO and D testing and DAT were performed using
commercially available reagents according to manufactur-
ers protocols using tube method (Immucor Gamma Inc.,
Norcross, GA). Antibody detection and identifcation were
performed using either gel technology according to the man-
ufacturers protocol using reagent cells (RBCs 0.8% Surgis-
creen and reagent RBCs 0.8% Resolve Antigram, ID-MTS
Gel Test; Ortho-Clinical Diagnostics, Raritan, NJ) or tube
method with commercially available cells and reagents
(PEG and LISS, Immucor Gamma Inc.). Titration studies
Table 1. Quantitative serologic monitoring
Approxi-
mate ges-
tational
age (wk)
Sero-
logic
fndings
Untreated
serum
titer
Untreated
serum
score
2-ME-
treated
serum
titer
2-ME-
treated
serum
score
7 Weak,
variable
reactivity
N/A N/A N/A N/A
14 Anti-Vel 4 24 4 24
18 Anti-Vel 4 27 4 14
22 Anti-Vel 8 N/T N/T N/T
26 Anti-Vel 16 25 4 18
34 Anti-Vel 16 30 4 14
N/A = not applicable; N/T = not tested.
Role for anti-Vel prenatal monitoring
W.J. Linz et al.
the bilirubin level after birth remains uncertain, although
liver immaturity is the most likely explanation. Finally, the
abnormal Doppler velocimetry in the absence of true fetal
anemia refects the inherent limitations of a screening test
that is known to have a 12 percent rate of false-positive re-
sults.
10
The Vel antigen (collection) and anti-Vel remain enig-
matic and are in need of further study. To date, it is known
that the Vel-negative phenotype is universally uncommon.
The incidence of the Vel-negative phenotype in the Hispanic
population is not known with certainty. Anti-Vel is famous-
ly associated with in vitro hemolysis, severe transfusion re-
actions, and shortened RBC survival,
11
although there is at
least one case report of Vel+ blood being successfully trans-
fused to a recipient possessing anti-Vel.
12
An auto-anti-Vel
13

has been described, although it is uncertain to what degree
the autoantibody contributed to the patients RBC destruc-
tion. Anti-Vel can be diffcult to detect, and it can simulate
the serologic profle of a cold-reactive autoantibody, with
disastrous consequences if improper analytic methods are
used.
14
The Vel antigen appears to have a variable antigenic
expression and appears to be best expressed on adult cells.
When viewed in concert with the recent case report by
van Gammeren et al.,
5
our case report further supports the
need not only to properly identify anti-Vel in the prenatal
setting but also to perform serial quantitative serologic
monitoring using methods to clearly separate the IgG frac-
tion (DTT treatment or 2-ME treatment) from the untreat-
ed (IgM and IgG) fraction. Although both the untreated
and treated serum titer and score results may be reported,
changes in the treated serum fraction, representing the IgG
component, may be a better trigger to initiate additional
clinical investigation. Additional cases may help identify
the most appropriate critical 2-ME or DTT titer for anti-
Vel.
References
Sussman LN, Miller EB. New blood factor: Vel. Rev He- 1.
matol 1952;7:36871.
Levine P, Robinson EA, Herrington LB, Sussman LN. 2.
Second example of the antibody for the high-incidence
blood factor Vel. Am J Clin Pathol 1955;25:7514.
Issitt PD, Oyen R, Reihart JK, Adebahr ME, Allen 3.
FH, Kuhns WJ. Anti-Vel 2, a new antibody show-
ing heterogeneity of Vel system antibodies. Vox Sang
1968;15:12532.
Tarsa M, Kelly TF. Managing a pregnancy with anti- 4.
bodies: a clinicians perspective. Immunohematology
2008;24:527.
van Gammeren AJ, Overbeeke MA, Idema RN, van 5.
Beek RH, Ten Kate-Booij MJ, Ermens AA. Haemolytic
disease of the newborn because of rare anti-Vel. Trans-
fus Med 2008;18:1978.
Roback JD. Technical manual. 16th ed. Bethesda, MD: 6.
AABB 2008: 900.
Marsh WL. Scoring of hemagglutination reactions. 7.
Judd WJ; Scientifc Section Coordinating Commit- 8.
tee of the AABB. Practice guidelines for prenatal and
perinatal immunohematology, revisited. Transfusion
2001;41:144552.
Reid ME, Lomas-Francis C. The blood group antigen 9.
factsbook. 2nd ed. San Diego, CA: Academia Press,
2004:503-4.
Giancarlo M. Noninvasive diagnosis by Doppler ultra- 10.
sonography of fetal anemia due to maternal red cell al-
loimmunization. N Engl J Med 2000;42:914.
Tilley CA, Crookston MC, Haddad SA, Shumak KN. Red 11.
blood cell survival studies in patients with anti-Ch
a
,
anti-Yk
a
, anti-Ge and anti-Vel. Transfusion 1977;17:169
72.
Issitt PD, Anstee DJ. Applied blood group serology. 4th 12.
ed. Durham, NC: Montgomery Scientifc Publications,
1998: 802.
Becton DL, Kinney TR. An infant girl with severe auto- 13.
immune hemolytic anemia: apparent anti-Vel specifc-
ity. Vox Sang 1986;51:10811.
Storry JR, Mallory D. Misidentifcation of anti-Vel due 14.
to inappropriate use of prewarming and adsorption
techniques. Immunohematology 1994;10:836.
Walter J. Linz, MD, MBA, Medical Director, Scott and
White Transfusion Service and Donor Center, Assistant
Professor of Pathology, Texas A&M UHSC, 2401 South 31st
Street, Temple, TX 76508; Judith T. Fueger, MT(ASCP)SBB,
Immunohematology Reference Laboratory, Diagnostic
Laboratories, BloodCenter of Wisconsin, Milwaukee, WI;
Steven Allen, MD, Associate Professor, Texas A&M UHSC/
Scott and White Memorial Hospital, Chair, Department
of Ob/Gyn, Director, Division of Obstetrics and Gynecol-
ogy, Scott & White Healthcare and Texas A&M University
Health Science Center College of Medicine, Temple, TX;
and Susan T. Johnson, MSTM, MT(ASCP)SBB, Immuno-
hematology Reference Laboratory, Diagnostic Laborato-
ries, BloodCenter of Wisconsin, Milwaukee, WI.
Notice to Readers
Immunohematology, Journal of Blood Group
Serology and Education, is printed on acid-free
paper.
Review
Granulocyte serology: current concepts and
clinical signifcance
M.E. Clay, R.M. Schuller, and G.J. Bachowski
Applying serologic procedures to the detection of RBC and lym-
phocyte antigens has facilitated the identification of granulo-
cyte antigens with established clinical significance, which are
now classified in the human neutrophil antigen system. Granu-
locyte alloantibodies and autoantibodies have been implicated
in a variety of clinical conditions including alloimmune neutro-
penia, autoimmune neutropenia, febrile and severe pulmonary
transfusion reactions, drug-induced neutropenia, refractori-
ness to granulocyte transfusions, and immune neutropenia
after hematopoietic stem cell transplantation. Although the
intrinsically fragile nature of granulocytes contributes to the in-
herent challenges of granulocyte serology, several advances in
laboratory procedures have improved detection of granulocyte
antibodies. This review will provide a current perspective about
the importance and use of granulocyte serology for detection
of granulocyte antibodies that have significant medical effects.
Key Words: granulocyte, neutrophil, antigens, antibodies, HNA,
alloimmune neonatal neutropenia, autoimmune neutropenia,
transfusion-related acute lung injury, TRALI
Clinical Signifcance
Alloimmune Neonatal Neutropenia
Neutropenia observed in the neonate is primarily the
result of neutrophil-specifc alloantibodies transplacentally
transferred to the fetus from the maternal circulation. The
presence of these human neutrophil antigen (HNA) anti-
bodies most likely is the result of prenatal maternal sensi-
tization by fetal neutrophils crossing the placental barrier,
although the antibodies may have also arisen in association
with autoimmune diseases such as systemic lupus
erythematosus or rheumatoi d arthri ti s.
2, 3
Mater-
nal al l oi mmunization can occur anytime after the frst
trimester of pregnancy. Once present in the fetal circula-
tion, maternal HNA antibodies will bind to the mature fetal
cells expressing the corresponding neutrophil alloantigens,
as the neutrophil precursor cells are spared.
4
Neonates af-
fected by alloimmune neonatal neutropenia (ANN) are al-
most always neutropenic at birth, although cases have been
reported in which the neutropenia has been delayed by 1 to
3 days.
4
ANN is often asymptomatic and goes undiscovered
unless the defcit of the neutrophils is profound enough
(absolute neutrophil count < 1.5 10
9
/L) that the newborn
becomes susceptible and succumbs to infection caused by
bacterial or fungal pathogens. These children commonly
present with fever, malaise, lethargy, skin infections such
as cellulitis and omphalitis, mucosal and respiratory infec-
tions (stomatitis, otitis media, and pneumonia), urinary
infections, and very rarely septicemia. These infections are
usually mild, but fatalities have been reported in 5 percent
of cases.
5
As there are numerous reasons for neonatal neu-
tropenia (i.e., congenital, antibody-mediated destruction,
or infection that decreases hemopoietic cell production),
detection of these maternal neutrophil antibodies is very
important for determining the mechanism causing neutro-
penia. This allows the physician to focus on the appropriate
treatment. The presence of maternal antibodies can persist
up to 6 months after birth although antibodies and clini-
cal effects usually dissipate more quickly. The incidence of
ANN has been estimated to be 0.1 to 0.2 percent.
6,7
Cur-
rently, neutrophil antibodies with the following specifcities
have been implicated in cases of ANN: HNA-1a,
810
HNA-
1b,
11
HNA-1c,
12
FcRIIIb (CD16),
13,14
HNA-2a,
15
HNA-3a,
16

and HNA-4a.
17
It has now been half a century since Lalezari
1
described the
frst case of neonatal neutropenia attributable to
transplacental transfer of granulocyte-specifc antibod-
ies, thus initiating the feld of granulocyte serology. This
event fostered several decades of work directed at the iden-
tifcation of granulocyte-specifc antigens (and antibodies),
methods for their detection, and an appreciation for their
clinical signifcance. In many ways, granulocyte serology
developed along the laboratory footprints of RBC and hu-
man lymphocyte antigen (HLA) serology with one major
exceptiongranulocytes could not be preserved for testing
purposes. The requirement for fresh cells was and contin-
ues to be a major impediment for wide-scale implementa-
tion of granulocyte immunobiology studies. Although the
number of laboratories worldwide that specialize in this
technology is limited, the clinical signifcance of granulo-
cyte antibodies has not diminished and is now supporting
the development of new methodologies that have the poten-
tial to enhance the utility of testing and growth of this feld.
Therefore, as we move ahead, it is important to (1) focus
on the clinical events that support the need for granulocyte
serology, (2) understand the benefts and limitations of the
current serologic assays, and (3) evaluate the application of
new technologies for the detection of granulocyte antigens
and antibodies.
Autoimmune Neutropenia
Autoimmune neutropenia (AIN) occurs in both in-
fants and adults. It is the most common form of chronic
neutropenia in infants. Severe neutropenia is usually recog-
nized 4 to 7 months after birth.
4
In the vast majority of AIN
cases infants exhibit relatively mild clinical issues such as
stomatitis, otitis, and respiratory infections. The presence
of clinical symptoms lessens as the child becomes older.
Neutropenia in infants is self-limited with complete reso-
lution commonly observed within 7 to 24 months.
18
How-
ever, this condition necessitates aggressive hygienic care of
infants to prevent infection and vigorous therapy that may
include severe antibiotic therapy along with treatment with
recombinant G-CSF and IVIG when infection occurs. Cell
destruction can occur in the peripheral circulation in addi-
tion to interference with myelopoiesis in the bone marrow.
Complete blood counts demonstrate absolute neutropenia,
although random fuctuations in the neutrophil count can
range from zero to near normal.
AIN is also observed in adults. Monocytosis with or
without lymphopenia may also be present.
19
As in infants,
the bone marrow commonly demonstrates an absence of
mature cells with an increase in myeloid precursors. This
apparent imbalance of precursors to mature segmented
neutrophils is distinct in cause from the maturation arrest
that occurs in congenital defects of neutrophils or hemato-
logic malignancy. Unlike RBC autoantibodies, antibodies in
AIN are often reported to have specifcity that can include
HNA-1a, HNA-1b, and HNA-2a.
1820
Drug-Induced Neutropenia
Although neutrophil antibodies are believed to be in-
volved in drug-induced neutropenia, neither the precise
mechanisms nor the particular antigens on the cell surface
have yet been clarifed. Some cases of drug-induced granu-
locytopenias probably result from direct marrow toxicity,
although immune-mediated processes also occur. Quinine,
quinidine, ibuprofen, and psychotropic medications such
as clozapine are recognized as potential causes of neutrope-
nia.
2123
Drug-dependent antibodies have been found to re-
act with both granulocytes and granulocyte precursors. The
FcRIIIb and CD177 neutrophil glycoproteins have been
reported to form neoantigens with drugs or their metabo-
lites, which are recognized by drug-dependent neutrophil
antibodies.
Transfusion Reactions
Antibodies to neutrophil and HLA antigens can result
in a variety of transfusion reactions. The spectrum of sever-
ity can range from mild febrile reactions to death in the case
of severe transfusion-related acute lung injury (TRALI). Re-
cently TRALI has been shown to be the most common cause
of transfusion-related fatalities in the United States.
24
Febrile nonhemolytic transfusion reactions are com-
mon and can occur when blood products containing WBCs
are transfused into patients who have leukocyte antibodies.
25

The occurrence of febrile reactions has been mitigated with
the use of leukocyte-reduced RBCs and platelet products.
As early as 1951 blood transfusions were implicated in
noncardiogenic lung edema.
26
The term TRALI was con-
ceived in 1985 when Popovsky and Moore
27
investigated a
series of 36 patients with well-defned transfusion-related
acute lung injury. They detected leukocyte (HLA and HNA)
antibodies in 89 percent of the implicated donors. Seventy-
two percent of these patients were treated with mechanical
ventilation, and 6 percent died. The leading theories for the
cause of TRALI involve the priming or activation of neu-
trophils with antigen-antibody interactions, inducing an
acute infammatory response.
28
Antibodies to HLA class I
and class II as well as HNAs have all been implicated in the
development of TRALI.
28
Even though HLA class I alloantibodies are the most
frequently encountered WBC antibodies in implicated blood
donors, they may act as much weaker triggers of acute lung
injury when compared with HLA class II and HNA antibod-
ies. In a study designed to investigate leukocyte antibody
specifcities in severe TRALI reactions, Reil et al.
29
reported
that even though 73 percent of leukocyte antibodies identi-
fed in their donor population were HLA class I, HLA class
II and HNA antibodies were associated with 81 percent of
the antibody-mediated TRALI cases.
29
Specifcally, in their
36 reported cases, 17 were elicited by blood products con-
taining HLA class II antibodies, and HNA antibodies were
implicated in 10 cases. Of the 10 fatal outcomes, 6 were
linked to HNA-3a antibodies in the transfused blood prod-
ucts, and HLA class II antibodies were associated with 3 fa-
talities.
Although leukocyte agglutinating (aggregating) anti-
bodies have been frequently associated with TRALI,
3034

the ability of HNA-3a antibodies to also prime and activate
neutrophils in the pulmonary vasculature is thought to be
central to severe and fatal TRALI.
35
Studies have now shown
that in addition to HNA-3a antibodies, HNA-2a and HNA-
4a antibodies also have the ability to prime neutrophils in
vitro.
3638
Although the pathophysiology of TRALI appears com-
plex and remains under investigation,
39
substantial clinical
and scientifc information suggests that HLA (especially
class II) and HNA antibodies do play a signifcant role in
this disorder.
28
Efforts are now being focused on detecting
HLA and HNA antibodies in donors both to retrospectively
investigate the cause of acute lung injury in blood recipients
and to serve as a valuable strategy in preventing TRALI by
identifying donors who may be responsible for the transfu-
sion reaction.
Granulocyte Antigens
Granulocyte-specifc antigens are those with a tissue
distribution restricted to granulocytes (neutrophils, eosino-
phils, and basophils), whereas neutrophil-specifc antigens
are only present on neutrophils. Because of the diffculty in
characterizing antigens on basophils and eosinophils, many
M.E. Clay et al.
Granuloycte serology: concepts and signifcance
antigens described as neutrophil-specifc have not been
tested to determine whether they are present only on neu-
trophils or on all granulocytes. Although granulocyte (neu-
trophil) antigen systems were initially identifed through
studies of ANN and AIN, both terms, granulocyte and neu-
trophil, have been used to describe the alloantigens.
Nomenclature
Lalezari and colleagues
8
identifed the frst granulo-
cyte antigen during their investigation of a neonate with
transient ANN and subsequently proposed the frst no-
menclature system for these antigens. Because their stud-
ies indicated that the antigen was neutrophil-specifc, they
designated it the N system and the antigens were named in
chronologic order of discovery. The antigens were labeled
alphabetically and the alleles were described numerically.
The system eventually classifed seven antigens: NA1,
8

NA2,
40
NB1,
41
NB2,
42,43
NC1,
44
ND1,
45
and NE1,
46
which be-
came the foundation for granulocyte serology.
During this same period and subsequently, several dif-
ferent granulocyte antigens were identifed. Some were
specifc for granulocytes (or neutrophils): HGA-3, GA,
GB, GC, Gr1, Gr2
4749
; some were shared with other cells
or tissues: HGA-1, 5a, 5b, 9, MART
47,5052
; and some had
unclassifed distribution: CN1, KEN, LAN, LEA, SL.
5356

All of these granulocyte antigens were detected with sera
from patients with autoantibodies or alloantibodies di-
rected against granulocytes, using traditional blood group
serologic techniques such as agglutination, immunofuo-
rescence, and cytotoxicity. Unfortunately, most of these
antigens have not been further classifed owing to the lack
of suffcient biologic material for study.
Although this led to a consolidation of
individuals working in this feld, advances
were made and efforts were directed at im-
munobiochemical, genetic, and structural
or functional studies for a few of the well-
classifed granulocyte antigens. With the
advent of this information came the need
for a standardized granulocyte antigen no-
menclature system based on the molecules
carrying the antigens and the genes encod-
ing each allele. In 1999, the Granulocyte
Antigen Working Party of the International
Society of Blood Transfusion (ISBT) estab-
lished a new nomenclature system for the
well-characterized granulocyte antigens,
which was based on the glycoprotein loca-
tion of the antigens.
57
In this system the
granulocyte antigens are called human neu-
trophil alloantigens to indicate they are ex-
pressed on neutrophils; however, this does
not imply that they are neutrophil-specifc.
Each antigen group is assigned a number,
and polymorphisms within the group are
designated alphabetically in sequential order
of detection. The HNA system currently includes seven an-
tigens, which are restricted to fve antigen groups. The key
features of the HNA system are summarized in Table 1. An
extensive amount of information is now known about these
antigens, and the reader is referred to Buxs recent and
comprehensive review about the antigens that constitute
this system.
58
HNA-1 is the best-characterized group, and it contains
three antigens: HNA-1a, HNA-1b, and HNA-1c. The HNA-1
antigens are expressed only on neutrophils, and antibodies
to these antigens are frequently implicated in cases of al-
loimmune or autoimmune neutropenia.
18,58
The antigens are
epitopes that occur on the neutrophil low-affnity FcRIIIb
receptor (CD16). This molecule binds the Fc region of IgG
antibodies complexed to other antigens or immunoglobu-
lins. The gene frequencies of HNA-1a, HNA-1b, and HNA-
1c vary among different ethnic populations.
58
Furthermore,
individuals lacking FcRIIIb have been identifed. These in-
dividuals do not express the HNA-1 antigens on their neu-
trophils, thus presenting as an HNA-1 null phenotype.
54,62
The HNA-2 group has one characterized antigen
HNA-2awhich is located on a 58- to 64-kDa glycoprotein
that is expressed only on neutrophils.
63
Although HNA-2a is
a high-frequency antigen phenotype (i.e., greater than 90%
for most populations), it is unique in that it is expressed
on subpopulations of neutrophils among antigen-positive
people. The expression of HNA-2a is greater on neutrophils
from women than men, and two or three subpopulations
of expression are often detected: one population that lacks
HNA-2a expression and one or two populations that express
the antigen but with different intensities.
64
The functional
Table 1. Human neutrophil antigens
Antigen
groups Antigens
Former
name
Antigen location/
glycoprotein CD
number
Coding gene
alleles
Clinical
signifcance
HNA-1 HNA-1a
HNA-1b
HNA-1c
NA1
NA2
SH
FcRIIIb/CD16b
FcRIIIb/CD16b
FcRIIIb/CD16b
FCGR3B*01
FCGR3B*02
FCGR3B*03
ANN, AIN,
TRALI,
HNA-2 HNA-2a NB1 5864 kDa/CD177 Unknown ANN, AIN, TRALI,
febrile transfu-
sion reactions,
drug-dependent
neutropenia
HNA-3 HNA-3a
HNA-3b
5b
5a
CTL2/Unknown
CTL2/Unknown
ANN, TRALI ,
febrile transfusion
reactions
HNA-4 HNA-4a MART CR3/CD11b
( M subunit)
ITGAM*01
(230G)
ANN
HNA-5 HNA-5a OND LFA-1/CD11a
( L subunit)
ITGAL*01
(2372G)
Unknown
Information from references 20, 58, 59, 60.
Alleles of the coding genes are named according to the ISGN Guidelines for Hu-
man Gene Nomenclature.
61
Except HNA-1c.
AIN = autoimmune neutropenia; ANN = alloimmune neonatal neutropenia; CD =
clusters of differentiation; CR3 = complement receptor 3; CTL2 = choline trans-
porter-like protein 2; FcRIIIb = Fc gamma receptor IIIb; HNA = human neutrophil
antigen; LFA-1 = leukocyte function antigen-1; TRALI = transfusion-related acute
lung injury.
role of HNA-2a is not known, and antibodies to this antigen
have been associated with AIN,
20
ANN,
65
febrile transfusion
reactions,
66
TRALI,
67
and drug-dependent neutropenia.
68
HNA-3a is expressed on granulocytes, lymphocytes,
platelets, endothelial cells, kidney, spleen, and placental
cells.
50
It is a high-frequency antigen located on choline
transporter-like protein 2 (CTL2), and its function is not
known.
59,60
Alloantibodies to HNA-3a have been associated
with occasional cases of febrile transfusion reactions
69
and
one case of ANN.
70
Although antibodies to HNA-3a are rare,
their major clinical signifcance has been their association
with serious and fatal TRALI events.
3234
The HNA-4 and HNA-5 antigens are located on subunits
of the 2 integrin family (CD11a and CD11b molecules). In-
tegrins are a large family of adhesive receptors that are es-
sential for cell-cell interactions and cell traffcking. HNA-4a
is an epitope on the M (CD11b) chain of CD11b/CD18 (the
CR3 molecule), and HNA-5a is an epitope on the L chain
(CD11a) of CD11a/CD18 (the leukocyte function antigen-1
[LFA-1] molecule).
71
HNA-4a is expressed on granulocytes,
monocytes, and natural killer cells, whereas HNA-5a is ex-
pressed on all leukocytes.
58
Alloantibodies to HNA-4a can
cause ANN,
72
and the CD11b/CD18 complex has been re-
ported to be the target of autoantibodies.
18,73
To date, HNA-
5a antibodies have not been clinically associated with
neutropenia.
A general consensus exists that granulocytes do not ex-
press ABO antigens
74
or HLA class II antigens.
75
Although
the density of HLA class I antigens on granulocytes is fairly
low, it can vary among individuals.
76
Therefore, these anti-
gens must be taken into consideration when patient samples
are tested for granulocyte antibodies. HLA class I antigens
may be of particular interest during granulocyte transfusion
for neutropenic patients as these antigens, in granulocyte
concentrates, may be introduced in high amounts to recipi-
ents with preformed class I antibodies. Transfusion reac-
tions including TRALI have been observed in this situation,
and the antigen-antibody interaction has been proposed to
limit effectiveness of the transfusion.
7779
Serologic Testing
The detection of HNA antibodies is labor intensive and
technically challenging. Unlike HLA testing, commercial
test kits are not readily available for the detection of granu-
locyte antibodies by solid-phase methodologies. Therefore,
intact viable granulocytes are required for granulocyte an-
tibody screening. Granulocytes are intrinsically designed
to respond to physiologic priming signals, so they must be
handled very carefully to prevent activation. Once activated
it is impossible to interpret any serologic test results involv-
ing these cells, as false-positive test results abound. Granu-
locytes are also extremely labile and must be used within
24 hours of collection, necessitating ready access to panel
cell donors with the needed antigen phenotypes. Although
many have tried, attempts to develop short-term or long-
term granulocyte preservation procedures have been un-
successful, thus requiring that fresh suspensions of granu-
locytes be prepared daily for testing procedures. Finally, the
presence of HLA antibodies in test sera makes identifying
HNA antibodies diffcult because granulocytes also express
HLA class I antigens on their cell surface.
Preparation of Pure Granulocyte Suspensions
The isolation of pure granulocyte suspensions from pe-
ripheral blood was greatly facilitated and simplifed with
Byums introduction in 1968 of the Ficoll-Isopaque gradi-
ent technique.
80
Double-density gradient centrifugation is
now the most widely used approach for the isolation and
purifcation of granulocytes for serologic testing. The proce-
dure makes use of discontinuous Ficoll gradients, and nu-
merous publications exist describing the methodology and
the recovery, purity, functional integrity, and activity of the
isolated granulocytes.
8184
Granulocyte Antibody Detection and Antigen Typing
Laboratory approaches for the detection of granulocyte
antibodies require procedures that are accurate, reproduc-
ible, and practical. Although numerous granulocyte sero-
logic assays have been described, not all meet these criteria.
In addition to the assay used, the ability to detect antibodies
in test specimens or antigens on test cells may also depend
on target cell antigen density and the concentration and
immunoglobulin type of the antibody in the test reagent.
Therefore, the successful detection of granulocyte antibod-
ies often requires that a combination of methods be used.
Although several methods for the detection of gran-
ulocyte-specifc antibodies are used by the few laborato-
ries that perform this testing, the ISBT Working Party on
Granulocyte Immunobiology, a worldwide consortium of
16 laboratories that participate in the annual International
Granulocyte Immunology Workshop (IGIW), have recom-
mended that granulocyte antibodies should be investigated
using a minimum of two methods: the granulocyte immun-
ofuorescence test (GIFT) and the granulocyte agglutination
test (GAT).
85
Multiple workshops have shown the sensitiv-
ity of GIFT to far exceed that of GAT; however, agglutinat-
ing antibodies such as HNA-1c, HNA-3a, and HNA-4a are
much more readily detectable in GAT and in many cases it
is the only reliable method for detecting these specifcities.
For example, the seventh IGIW included a test specimen
that contained an HNA-3a antibody that could only be de-
tected by GAT and the granulocyte chemiluminescence test
(GCLT). Although GCLT has been useful in selective labora-
tory settings, it is not routinely used for clinical granulocyte
antibody detection and is not discussed.
M.E. Clay et al.
Antibody Detection Methods
Granulocyte Agglutination Test
Lalezari
86
developed the basis of this assay in the early
1960s. Since that time, micro techniques and the use of
pure granulocyte suspensions with optimized concentra-
tions of EDTA have vastly improved the performance of
this test. Typically, a panel of three to fve donors is selected
to include all known neutrophil antigens currently defned
by the ISBT consortium. Purifed granulocytes are isolated
from EDTA-anticoagulated peripheral whole blood using
a double-density gradient. These granulocytes are then
washed in PBS before use. The test is biphasic and is based
on the intrinsic response of granulocytes to aggregate when
stimulated by antibodies reacting to corresponding cell
surface antigens. The resulting agglutination is the conse-
quence of the granulocyte activation that occurs during the
sensitization phase, which induces the cells to form pseu-
dopods and slowly migrate toward one another during the
aggregation phase, until membrane contact is established.
Typically, the serum or plasma from the specimen being
investigated is incubated with the granulocyte suspension
for 4.5 to 6 hours at 30C.
87
The reactions are evaluated us-
ing an inverted-phase microscope and graded from nega-
tive to 4+ on the basis of the percentage of cells that are
agglutinated. Our laboratory has determined cutoff scores
for both adult and pediatric patients (<6 years old) on the
basis of the reaction grades established during microscopic
evaluation. Both IgG and IgM antibodies are detected by
this method.
Granulocyte Immunofuorescence Test
Developed by Verheugt et al.
88
in the late 1970s, this
fuorescent antiglobulin technique is used for the detec-
tion of granulocyte alloantibodies and autoantibodies that
are circulating and cell bound. As with the GAT, an anti-
body screening panel of three to fve donors is selected
to include all currently defned HNA antigens. A purifed
granulocyte suspension is prepared and treated with 1%
paraformaldehyde for a short time to prevent nonspecifc
immunoglobulin binding to the neutrophil Fc receptors and
to stabilize the cell membrane. Serum or plasma from the
patient is then incubated with an optimized concentration
of granulocytes for 30 minutes at 37C. After a wash step
that removes unbound antibodies, the granulocytes are
then incubated with F(ab)
2
fragments of a fuorescent con-
jugated anti-human antibody for approximately 30 min-
utes at room temperature in a dark environment. The as-
says performance is optimized with the use of a fuorescent
secondary probe that can detect both IgM and IgG antibod-
ies (to ensure the detection of both primary and secondary
immune responses) and the use of F(ab)
2
Ig fragments to
prevent the probe from binding to the high concentrations
of Fc receptors on granulocyte surface membranes. The cells
undergo another wash cycle, are resuspended, and then are
analyzed. Detection of the immunofuorescence reactions
can be accomplished by using either a fuorescence-detect-
ing microscope or a fow cytometer. Evaluation by a fow
cytometer has in most cases replaced the fuorescence mi-
croscope as more cells can be analyzed in a shorter time,
resulting in improved assay sensitivity and reproducibility,
and it does not require the expertise needed to evaluate the
characteristic staining pattern seen with specifc granulo-
cyte antibodies by microscopic analysis.
Monoclonal Antibody Immobilization of Granulocyte Antigens
The monoclonal antibody immobilization of granulo-
cyte antigens assay (MAIGA) is based on the platelet ver-
sion (MAIPA) developed by Kiefel et al. in 1987.
89
MAIGA
relies on the capture of neutrophil-specifc antigen-antibody
complexes by a murine monoclonal antibody onto a solid-
phase surface.
90
The benefts of this test are twofold: frst,
this is currently the most sensitive assay for the detection of
granulocyte antibodies, and second, the assay is designed to
detect only HNA antibodies even when HLA antibodies are
present in the test specimen. The disadvantage of this pro-
cedure is that it is very complex and requires highly skilled
staff to perform the detailed techniques.
Granulocytes are incubated with the test serum or plas-
ma for 30 minutes at 37C. This granulocyte suspension is
then washed to remove any unbound immunoglobulins and
incubated with a murine monoclonal antibody to a specifc
neutrophil glycoprotein for an additional 30 minutes. Af-
ter another wash step, the granulocyte membranes are dis-
rupted in a mild detergent and centrifuged. The resulting
lysate is then transferred to polystyrene microwells coated
with anti-mouse immunoglobulins for incubation. The
trimolecular neutrophil antigenpatient HNA antibody
murine monoclonal antibody complex present in the lysate
is captured on the solid phase, whereas any HLA
antibody-antigen complexes (if present) are removed by
a subsequent wash step. Remaining complexes are then
detected by the addition of anti-human IgG conjugated to
horseradish peroxidase followed by a substrate (OPD dis-
solved in 30% H
2
O
2
), and the reaction is analyzed with a
spectrophotometer.
MAIGA can be used to detect antibodies specifc
to HNA-1a, HNA-1b, and HNA-1c antigens located on
FcRIIIb (CD16); HNA-2a present on gp 58- to 64-kDa
(CD177); HNA-4a on the complement component receptor
C3bi (CD11b); and HNA-5a on the LFA-1 receptor (CD11a).
HNA-4a and HNA-5a antibodies can also be detected by a
common CD18 monoclonal antibody.
Antigen Detection and Typing Methods
The same methodology that is used to detect antibodies
in serum and plasma can be used to characterize the antigen
phenotype of an individuals granulocytes. Just as the short
lifespan of granulocytes is a limitation to obtaining panel
cells for antibody testing, it also restricts the ability to do
antigen typing because granulocyte specimens must be tested
within 24 hours of collection to obtain reliable results. A
suffcient volume of whole blood must be collected to obtain
an adequate number of granulocytes for phenotyping. This
can be a dilemma in pediatric patients or patients who are
neutropenic. Because ample volume, freshness, and avail-
ability of granulocytes is critical, this also acts as a practical
barrier to doing donor-recipient crossmatch testing during
investigations of suspected granulocyte-antibodymediat-
ed transfusion reactions such as TRALI. The highly charac-
terized antisera used to phenotype granulocytes should be
free of HLA class I antibodies as their presence can result in
a false-positive HNA typing. At a minimum, the titer of the
HLA antibody should be weaker than the HNA antibody
as it can then be readily diluted to a nonreactive level. An-
other constraint to phenotyping granulocytes is the limited
availability of qualifed antisera for the currently defned
HNA antigens.
Monoclonal antibodies specifc to several HNA anti-
gens are commercially available and have been used to type
granulocytes by GIFT using the fow cytometer. An advan-
tage to phenotyping by this method is that it can be done
with whole blood instead of with isolated granulocytes.
Because phenotyping granulocytes cannot often be per-
formed or can be unreliable if samples are not handled in a
careful and timely manner, genotyping by PCR, which has
less stringent sample age and handling restraints, can serve
either as an alternative to or as confrmation of serologic
results.
Granulocyte Genotyping
The discovery of the PCR in the 1980s facilitated the
development of molecular methods that have now been ap-
plied to delineate the genetics and allelic variations of the
HNAs. Initially, the characterization of the genes encoding
the HNA-1 antigen system (HNA-1a, HNA-1b, and HNA-1c)
allowed for the development of PCR assays using sequence-
specifc primers (SSP) to differentiate the alleles. To com-
plicate matters there is a high degree of homology between
the FCGR3A gene that encodes the FcRIIIa granulocyte
receptor and the FCGR3B gene where the three different
HNA-1 polymorphisms reside. HNA-1a, HNA-1b, and HNA-
1c are encoded by FCGR3B*1, FCGR3B*2, and FCGR3B*3
genes, respectively. FCGR3B*1 differs from FCGR3B*2
by fve nucleotide bases, and a single nucleotide polymor-
phism (SNP) differentiates FCGR3B*2 from
FCGR3B*3 (Table 2). FCGR3A differs from
FCGR3B at fve different nucleotide locations
(Table 2).
Methods are also available to genotype
the HNA-4a and HNA-5a alleles.
71
Both of
these polymorphisms are the result of SNPs.
Even though the molecular sequence of HNA-
2a has been identifed, genotyping methods
to detect the gene that encodes this alloanti-
gen are not available. The HNA-2a negative
phenotype is attributable to CD177 mRNA
splicing defects, and no mutations have been detected in
the CD177 introns or exons in these individuals.
93
Because
the gene encoding HNA-3a has only recently been identi-
fed, there has not been a genotyping procedure for this
antigen, but the clinical importance of this antigen is sup-
porting efforts to get a molecular method developed.
Genomic DNA (gDNA) can be isolated and purifed
from anticoagulated blood using any number of published
methods. A unique set of sense and antisense oligonucle-
otide primers that border the DNA fragment to be amplifed
are added to a master mix of reagents that includes deoxy-
nucleoside triphosphates (dNTPs), Taq polymerase, a buf-
fer solution containing Mg
2+
, and the individuals gDNA.
The master mix is subjected to a series of 30 to 40 tempera-
ture changes that defne the amplifcation cycles. Each cycle
consists of three temperature steps that are critical in the
amplifcation of the DNA target strands. The frst step in
the cycle typically involves heating the master mix to a tem-
perature of 95C. This denaturation step results in double-
stranded DNA separating into single strands by disrupting
the hydrogen bonds between each nucleotide base pair. The
annealing step follows in which the reaction temperature is
lowered to 50C to 65C. This allows the primers to hybrid-
ize to the opposing strands of the target DNA. The fnal ex-
tension step heats the master mix from 71C to 72C, which
results in DNA synthesis in the presence of Taq polymerase
and dNTPs complementary to the oligonucleotide primers.
A fnal extension step is added after all cycles have been
completed to ensure that any remaining single-stranded
DNA is fully elongated. The amplifed double-stranded
DNA (amplicon) is separated by size using agarose gel
electrophoresis, stained with ethidium bromide, and visu-
alized by fuorescence using ultraviolet light. The sizes of
the amplicons are determined by comparison with a DNA
ladder, which contains DNA fragments of known sizes that
were run alongside the PCR products during the agarose gel
electrophoresis step.
Future Directions
Solid-Phase Technology
The complement-dependent lymphocyte cytotoxicity
assay was the frst platform, developed nearly 50 years ago,
for identifying HLA antibodies and remains the standard
for the development of new methodologies in this feld.
M.E. Clay et al.
Table 2. Nucleotide differences for the three HNA-1 alleles (1a, 1b, and 1c) on
FcRIIIb and among the FcRIII genes
HNA Gene Polymorphic nucleotides
141 147 227 266 277 349 473 505 559 641 733
1a FCGR3B*1 G C A C G G A C G C T
1b FCGR3B*2 C T G C A A A C G C T
1c FCGR3B*3 C T G A A A A C G C T
FCGR3A G C G C G A G T T T C
Information from references 91 and 92.
NOTE: Boldface indicates substituted nucleotides.
FcRIIIb = Fc gamma receptor IIIb; HNA = human neutrophil antigen.
Multiple new technologies are now being used by HLA
laboratories, and in the past few years solid-phase assays
have been developed in which purifed HLA molecules are
bound to a well in a microtiter plate or to a bead. HLA an-
tibody is then detected either using an enzyme-linked im-
munosorbent assay (ELISA) technique or by analyzing the
beads with a fow cytometer, which uses a new technology
developed by Luminex (Austin, TX).
94
A critical advantage
of this technology is its ability to measure multiple analytes
simultaneously in a single reaction system. The commer-
cially available test kits use purifed HLA antigens bound
to microbeads, which can be individually differentiated by
varying ratios of internal fuorescent dye prepared during
the manufacturing process. The test involves incubation of
test sera with the panel of microbeads, and the bound HLA
antibody is detected by a secondary R-phycoerythrin (PE)
conjugated IgG (or IgM)specifc antibody.
Analysis is done with the use of a Luminex fow analyz-
er in which the beads pass through two lasers, similarly to
classic fow cytometry. One laser identifes the specifc bead
being analyzed, thereby identifying the unique HLA anti-
gen. The other laser detects the presence of bound Ig on the
surface of the microbead. The color signals are then detect-
ed and processed into data for each reaction. A signifcant
fuorescence shift from the negative control range indicates
the presence of specifc antibody in the test sample.
The widespread use of Luminex technology by HLA
laboratories has fostered commercial activity directed at
using this technology for the detection of HNA antibodies
that have been related to febrile and pulmonary transfusion
reactions. One Lambda, Inc. (Canoga Park, CA), has devel-
oped recombinant cell lines for HNA-1a, -1b, -1c, -2a, and
-4a, and these cell lines have allowed the production of pu-
rifed proteins that have been individually immobilized on
Luminex microbeads for use in their LABScreen Multifow
bead assay. Stroncek and colleagues
95
evaluated the specifc-
ity and sensitivity of this new procedure using 22 sera that
had known HNA antibodies detected with the GA assay and
140 samples from nontransfused men. All HNA antibody
specifcities (except 4a) recognized by GAT were detected
in the solid-phase fow bead system (8 HNA-1a, 6 HNA-1b,
1 HNA-1c, and 8 HNA-2a positive sera), and HNA-1c reac-
tivity was detected in four specimens that were negative by
GAT. In addition, all of the 140 negative control samples
had negative reactions when tested by LABSreen Multi, and
the system was able to detect both HNA and HLA antibod-
ies (both class I and class II) in some individual specimens.
95

Although preliminary, these results suggest that this solid-
phase technology may emerge as a major achievement in
the development of a new sensitive and specifc platform
for the detection of granulocyte antibodies in an automated
manner with limited complexity and high cost effciency.
Cell Lines Expressing HNA Antigens
Another approach to eliminate the need for freshly
isolated granulocytes for granulocyte antibody screening
has been the establishment of cell lines that express HNA
antigens. In 1999 Buxs group transfected Chinese ham-
ster ovary (CHO) cells with HNA-1a, HNA-1b, and HNA-
1c cDNA and established stable cell lines expressing these
antigens.
96
Although the cell lines were stable for 1 month
at 4C, they demonstrated high background fuorescence in
the GIF assay and could only be used for the MAIGA, which
limited their overall utility. More recently, Yasui et al.
97,98

transduced HNA genes into the human erythroleukemia
cell line K562 and established a panel of K562 cell lines that
stably expressed HNA-1a, -1b, -1c, -2a, -4a, -4b, -5a, and -5b
for at least 1 year while being maintained in culture. They
were not able to establish a cell line expressing HNA-3a be-
cause of the previous lack of biochemical and genetic infor-
mation for this antigen. The groups report demonstrated
the utility of using the cultured cells in the GIF assay to de-
tect HNA antibodies in serum samples from patients with
HNA antibodyrelated disorders, but experience with this
new cell line technology is very limited and it remains to be
seen whether the cell lines can be maintained or preserved
in a manner that would support the use of these cells for
routine (nonresearch) granulocyte serology applications.
ISBT Working Party on Granulocyte Immunobiology
In 1989, Professor Alan Waters organized the UK Plate-
let and Granulocyte Serology Working Group, and 11 inter-
national laboratories participated in the First International
Workshop on Granulocyte Serology. The objectives of this
frst workshop were as follows: (1) to initiate an exchange
of sera among laboratories to evaluate the techniques be-
ing used for the detection of granulocyte antibodies, (2) to
foster working relationships among laboratories involved
in this feld, and (3) to support the development of the sci-
ence and technology for the detection of granulocyte anti-
bodies.
99
The Working Party was initially affliated with the Brit-
ish Blood Transfusion Society but is now affliated with the
ISBT. The goals of the Second Workshop, in 1996, were
the (1) establishment of typed granulocyte panels by well-
defned typing sera, (2) profciency testing of granulocyte
antibody detection, (3) investigation of uncharacterized
sera, (4) profciency testing of HNA genotyping, and (5) ex-
change of information to establish standards in granulocyte
serology.
100
Fifteen international laboratories participated
in the Fourth Workshop, in 2001, with the main objec-
tive being the establishment of a formal quality assurance
(QA) scheme for granulocyte serology and molecular typing
methods.
85
To date, the Working Party has conducted nine Granu-
locyte Serology Workshops and comprises 16 international
member laboratories, which are listed in Table 3. Also, the
American Red Cross Mid-America Blood Services Neutro-
phil & Platelet Immunology Laboratory has been a member
of this consortium since its inception in 1989. The workshops
continue to focus on the QA and standardization of the test
systems, and the Working Party recently published its
serologic screening recommendations for the investigation
and prevention of leukocyte antibody-mediated TRALI.
101
Conclusions
Granulocyte antibody and antigen testing has played
and continues to play a critical role for diagnosing and
investigating relatively rare but potentially lethal clini-
cal conditions such as TRALI and immune neutropenias.
Although mitigation strategies are already being imple-
mented nationwide, TRALI still appears to be a leading and
preventable cause of transfusion fatalities. As advancing
technology allows granulocyte antibody testing to be done
on a less manual and larger scale with more standardized
and reproducible results, it now has the potential to become
even more valuable as a tool to increase blood safety by ac-
tually preventing transfusion reactions without compro-
mising an adequate supply of blood.
Acknowledgments
The authors would like to thank Penny Milne and Bob-
bie Gibson for their helpful assistance with the preparation
of the manuscript.
References
Lalezari P, Nussbaum M, Gelman, et al. Neonatal 1.
neutropenia due to maternal isoimmunization. Blood
1960;15:23643.
Bux J, Jung KD, Kauth T, et al. Serological and clinical 2.
aspects of granulocyte antibodies leading to alloimmune
neonatal neutropenia. Transfus Med 1992;2:1439.
Rodwell RL, Gray PH, Taylor KM, et al. Granulocyte 3.
colony stimulating factor treatment for alloimmune
neonatal neutropenia. Arch Dis Child Fetal Neonatal
Ed 1996;75:F578.
Rodney GE, Lalezari P. HLA and neutrophil antigen 4.
and antibody systems. In: Hoffman R, Benz E, Shattil
SJ, eds. Hematology basic principles and practice. 3rd
ed. New York: Churchill Livingstone, 2000:222041.
Lalezari P. Alloimmune neonatal neutropenia. In: En- 5.
gelfriet CP,Van Logham JJ, von dem Borne AEGKr,
eds. Immunohaematology. Amsterdam: Elsevier Sci-
ence Publishers, 1984:17986.
Clay M, Kline W, McCullough J. Serological examina- 6.
tion of postpartum sera for the presence of neutrophil-
specifc antibodies (abstract). Transfusion 1981;21:616.
Levine DH, Madyastha P, Wade R, Levkoff AH. Neo- 7.
natal isoimmune neutropenia (abstract). Pediatr Res
1981;15:296A.
Lalezari P, Bernard GE. An isologous antigen-antibody 8.
reaction with human neutrophils, related to neonatal
neutropenia. J Clin Invest 1966;45:174150.
Verheugt FWA, van Noord-Bokhorst JC, von dem 9.
Borne AEGKr, et al. A family with alloimmune neonatal
neutropenia: group specifc pathogenicity of maternal
antibodies. Vox Sang 1979;36:18.
van der Weerdt CM, Lalezari P. Another example of 10.
isoimmune neonatal neutropenia due to anti-NA1. Vox
Sang 1972;22:43846.
Boxer LA, Yokoyama M, Lalezari P. Isoimmune neona- 11.
tal neutropenia. J Pediatr 1972;80:7837.
Bux J, Ernst-Ludwig Stein, Bierling P, et al. Character- 12.
ization of a new alloantigen (SH) on the human neutro-
phil Fc gamma receptor IIIb. Blood 1997;89:102734.
Schuller RM, Stroncek DF, Alpan O, et al. Different re- 13.
sponses observed between G-CSF and GM-CSF therapy
in a case of neonatal alloimmune neutropenia (ab-
stract). Transfusion 2008;48(Suppl):53A.
Bachowski GJ, Mair DC, Schuller RM, et al. Alloimmune 14.
neonatal neutropenia in monozygous twins caused by
antibody to FcRIIIb in a mother lacking CD16 antigen
(abstract). Transfusion 2007;47(Suppl):21A.
Stroncek DF, Sharpiro RS, Filipovich AH, et al. 15.
Prolonged neutropenia resulting from antibodies to
neutrophil-specifc antigen NB1 following marrow
transplantation. Transfusion 1993;33:15863.
M.E. Clay et al.
Table 3. Members of the ISBT Working Party on Granulocyte
Immunobiology
Platelet and Leucocyte Immunology Laboratory, EFS Ile de France,
Hopital Henri Mondor, Creteil, France
Leucocyte and Platelet Immunology Laboratory, Blood Service West of
the German Red Cross, Hagen, Germany
The BloodCenter of Wisconsin, Platelet and Neutrophil Immunology
Laboratory, Milwaukee, WI, USA
Institute of Transfusion Medicine, University of Schleswig-Holstein, Kiel,
Germany
Australian Red Cross Blood Service-Queensland, Brisbane, Australia
Platelet and Granulocyte Immunology Laboratory, National Blood Se
vice, Bristol, UK
Blood Transfusion Center of Slovenia, Ljubljana, Slovenia
Centre de Transfusio I Banc de Teixitis, Hospital Vall dHebron, P. Vall
dHebron, Barcelona, Spain
Sanguin Diagnostics, Immunohaematology Diagnostic Department,
Amsterdam, The Netherlands
Institute for Clinical Immunology and Transfusion Medicine, Justus
Liebig University, Giessen, Germany
American Red Cross Mid-America Blood Services, Neutrophil & Platelet
Immunology Laboratory, St. Paul, MN, USA
Department of Transfusion Medicine, Graduate School of Medical Sc
ences, University of Tokyo, Tokyo, Japan
Department of Immunohaematology and Transfusion Medicine, Institute
of Haematology and Blood Transfusion, Warsaw, Poland
Aberdeen and North East Scotland Blood Centre, Scottish National
Blood Transfusion Service, Aberdeen, UK
Platelet and Leucocyte Immunology Laboratory, Institute of Biology,
Nantes, France
Department of Transfusion Medicine, Karolinska University Hospital,
Stockholm, Sweden
de Haas M, Muniz-Diaz E, Alonso LG, et al. Neutrophil 16.
antigen 5b is carried by a protein, migrating from 70 to
95 kDa, and may be involved in neonatal alloimmune
neutropenia. Transfusion 2000;40:2227.
Fung L, Pitcher LA, Willett JE, et al. Alloimmune neo- 17.
natal neutropenia linked to anti-HNA-4a. Transfus
Med 2003;13:4952.
Bux J, Behrens G, Jaeger G, et al. Diagnosis and clinical 18.
course of autoimmune neutropenia in infancy: analysis
of 240 cases. Blood 1998;91:1816.
Bux J, Kissel K, Nowak K, et al. Autoimmune neutrope- 19.
nia: clinical and laboratory studies in 143 patients. Ann
Hematol 1991;63:24952.
Stroncek D. Granulocyte antigens and antibody detec- 20.
tion. Vox Sang 2004;87(Suppl 1):914.
Sperner-Unterweger B, Czeipek I, Gaggl S, Geissler 21.
D, Spiel G, Fleischhacker WW. Treatment of severe
clozapine-induced neutropenia with granulocyte col-
ony stimulating factor (G-CSF). Remission despite
continuous treatment with clozapine. Br J Psychiatry
1998;172:824.
Stroncek DF. Drug-induced immune neutropenia. 22.
Transfus Med Rev 1993;7:26874.
Mamus SW, Burton JD, Groat JD, et al. Ibuprofen- 23.
associated pure white cell aplasia. N Engl J Med
1986;314:6245.
Holness L, Knippen MA, Simmons L, et al. Fatalities 24.
caused by TRALI. Transfus Med Rev 2004;18:1848.
Brittingham TE. Immunologic studies on leukocytes. 25.
Vox Sang 1952;2:2428.
Barnard, RD. Indiscriminate transfusion: a critique of 26.
case reports illustrating hypersensitivity reactions. N Y
State J Med 1951;51:2399402.
Popovsky MA, Moore SB. Diagnostic and pathogenic 27.
considerations in transfusion-related acute lung injury.
Bux J, Sachs UJ. The pathogenesis of transfusion- 28.
related acute lung injury (TRALI). Br J Haematol
2007;136:78899.
Reil A, Keller-Stanislawski B, Gunay S, et al. Specifci- 29.
ties of leukocyte alloantibodies in transfusion-related
acute lung injury and results of leukocyte antibody
screening of blood donors. Vox Sang 2008;95:31317.
Ward HN. Pulmonary infltrates associated with leu- 30.
koagglutinin transfusion reactions. Ann Intern Med
1970;73:68994.
Thompson JS, Severson CD, Parmely MJ, et al. Pul- 31.
monary hypersensitivity reactions induced by trans-
fusion of non-HLA lekoagglutinins. N Engl J Med
1971;284:11205.
Nordhagen R, Conradi M, Drotorp SM. Pulmonary re- 32.
action associated with transfusion of plasma containing
anti-5b. Vox Sang 1986;51:1027.
Kopko PM, Marshall CS, MacKenzie MR, et al. Transfu- 33.
sion-related acute lung injury. Report of a clinical look-
back investigation. JAMA 2002;287:196871.
Davoren A, Curtis BR, Shulman IA, et al. TRALI due to 34.
granulocyte-agglutinating human neutrophil antigen-
3a (5b) alloantibodies in donor plasma: a report of 2
fatalities. Transfusion 2003;43:6415.
Silliman CC, Curtis BR, Kopko PM, et al. Donor anti- 35.
bodies to HNA-3a implicated in TRALI reactions prime
neutrophils and cause PMN-mediated damage to hu-
man pulmonary microvascular endothelial cells in a
two-event in vitro model. Blood 2007;109:17525.
Sachs UJ, Chavakis T, Fung L, et al. Human alloanti- 36.
body anti-MART interferes with Mac-1-dependent leu-
kocyte adhesion. Blood 2004;104:72734.
Kopko PM, Curtis BR, Kelher M, et al. Merging the 37.
pathogenesis of transfusion-related acute lung injury:
the priming activity of the 5b (HNA-3) antibody (ab-
stract). Transfusion 2004;44(Suppl 1):22A.
Sachs UJ, Hattar K, Weissman N, et al. Antibody- 38.
induced neutrophil activation as a trigger for trans-
fusion-related acute lung injury in an ex vivo rat lung
model. Blood 2006;107:12179.
Fung YL, Silliman CS. The role of neutrophils in the 39.
pathogenesis of transfusion-related acute lung injury.
Transfusion Med Rev 2009;23:26683.
Lalezari P, Radel E. Neutrophil-specifc antigens: im- 40.
munology and clinical signifcance. Semin Hematol
1974;11:28190.
Lalezari P, Murphy GB, Allen FH. NB1, a new neutro- 41.
phil antigen involved in the pathogenesis of neonatal
neutropenia. J Clin Invest 1971;50:110815.
Huang ST, Lin J, McGowan El, et al. NB2, a new allele 42.
of NB1 antigen involved in febrile transfusion reaction
(abstract). Transfusion 1982;22:426.
Lalezari P, Petrosova M, Jiang AF. NB2, an allele of 43.
NB1 neutrophil specifc antigen: relationship to 9a (ab-
stract). Transfusion 1982;22:433.
Lalezari P, Thelenfeld B, Weinstein WJ. The third neu- 44.
trophil antigen. In: Terasaki PI, ed. Histocompatibility
testing. Baltimore: Williams & Wilkins, 1970:31922.
Verheugt FWA, von dem Borne AEGKr, van Noord- 45.
Bokhorst JC, et al. ND1, a new neutrophil granulocyte
antigen. Vox Sang 1978;35:1317.
Claas FHJ, Langerak J, Sabbe LJM, et al. NE1, a 46.
new neutrophil specifc antigen. Tissue Antigens
1979;13:12934.
Thompson JS, Overlin VL, Herbick JM, et al. New 47.
granulocyte antigens demonstrated by microgranulo-
cytotoxicity assay. J Clin Invest 1980;65:14319.
Hasegawa T, Bergh OJ, Mickay MR, et al. Prelimi- 48.
nary human granulocyte specifcities. Transplant Pro
1975;7(Suppl 1):7580.
Caplan SN, Berkman EM, Babior BM. Cytotoxins 49.
against a granulocyte antigen system: detection by a
new method employing cytochalasin-B-treated cells.
Vox Sang 1977;33:20611.
Van Leeuwen A, Eerrise JG, van Room JJ. A new leu- 50.
kocyte group with two alleles: leukocyte group fve. Vox
Sang 1964;9:4317.
Van Rood JJ, van Leeuwen A, Schippers AMJ, et al. 51.
Immunogenetics of the group four, fve and nine sys-
tems. In: Curton ES, Mattiuz PI, Tosi RM, eds. Histo-
compatibility testing. Baltimore: Williams & Wilkins,
1967:20319.
Kline WE, Press C, Clay M, et al. Three sera defning a 52.
new granulocyte-monocyteT lymphocyte antigen. Vox
Sang 1986;50:1816.
Madyastha PR, Glassman A, Levine D, et al. Identif- 53.
cation of a new neutrophil antigen (CN1) more preva-
lent among American blacks (abstract). Transfusion
1983;23:426.
Schnell M, Halligan G, Herman J. A new granulocyte 54.
antibody directed at a high frequency antigen causing
neonatal alloimmune neutropenia (abstract). Transfu-
sion 1989;29(Suppl):46S.
Rodwell RL, Tudehope DI, ORegan P, et al. Alloimmune 55.
neonatal neutropenia in Australian aboriginals: an un-
recognized disorder? Transfusion Med 1991;1:637.
Stroncek DF, Ramsey G, Herr GP, et al. Identifcation of 56.
a new white cell antigen. Transfusion 1994;34:70611.
Bux J, Bierling P, von dem Borne AE, et al. ISBT Granu- 57.
locyte Antigen Working Party. Nomenclature of granu-
locyte alloantigens. (letter).Vox Sang 1999;77:251.
Bux J. Human neutrophil alloantigens. Vox Sang 58.
2008;94:27785.
Greinacher A, Wesche J, Hammer E, et al. Character- 59.
ization of the human neutrophil alloantigen-3a. Nature
Med 2010;16:458.
Curtis BR, Cox NJ, Sullivan MJ, et al. The neutrophil 60.
alloantigen HNA-3a (5b) is located on choline trans-
porter-like protein 2 (CTL2) and appears to be en-
coded by an R>Q 154 amino acid substitution. Blood
2010;115:2073-6.
Shows TB, McAlpine PJ, Boucheix C, et al. Guidelines 61.
for human gene nomenclature. An international system
for human gene nomenclature (ISGN, 1987). Cytogenet
Cell Genet 1987;46:1128.
Stroncek DF, Skubitz KM, Plachta LB, et al. Alloim- 62.
mune neonatal neutropenia due to an antibody to the
neutrophil Fc-gamma receptor III with maternal def-
ciency of CD16 antigen. Blood 1991;77:157280.
Stroncek DF, Skubitz K, McCullough JJ. Biochemical 63.
nature of the neutrophil-specifc antigen NB1. Blood
1990;75:74455.
Matsuo K, Lin A, Procter JL, et al. Variations in the 64.
expression of granulocyte antigen NB1. Transfusion
2000;40:65462.
Bux J, Jung KD, Kauth T, et al. Serological and clinical 65.
aspects of granulocyte antibodies leading to alloimmune
neonatal neutropenia. Transfusion Med 1992;2:1439.
Bux J. Granulocyte antibody mediated neutropenias 66.
and transfusion reactions. Infusions Ther Transfusion
Med 1999;26:1527.
Bux J, Becker F, Seeger W, et al. Transfusion-related 67.
acute lung injury due to HLR-A2-specifc antibodies in
recipient and NB1-specifc antibodies in donor blood.
Br J Haematol 1996;93:70713.
Stroncek DF, Shankar RH, Herr GP. Quinine- 68.
dependent antibodies to neutrophils react with a 60
kD glycoprotein on which neutrophil-specifc antigen
NB1 is located and an 85 kD glycosylphosphatidylinos-
itollinked N-glycosylated plasma membrane protein.
Blood 1993;81:275866.
Lalezari P, Bernard GE. Identifcation of a specifc 69.
leukocyte antigen: another presumed example of 5b.
de Haas M, Muniz-Diaz E, Alonso LG, et al. Neutrophil 70.
5b is carried by a protein, migrating from 70 to 95 kDa,
and may be involved in neonatal alloimmune neutrope-
nia. Transfusion 2000;40:2227.
Simsek S, van der Schoot CE, Daams M, et al. Molecu- 71.
lar characterization of antigenic polymorphisms (OND
a
and MART
a
) of the
2
family recognized by human leu-
kocyte alloantisera. Blood 1996;88:13508.
Fung L, Pitcher LA, Willett JE, et al. Alloimmune neo- 72.
natal neutropenia linked to anti-HNA-4a. Transfusion
Med 2003;13:4952.
Hartman KR, Wright DG. Identifcation of autoantibod- 73.
ies specifc for the neutrophil adhesion glycoproteins
CD11b/CD18 in patients with autoimmune neutro-
penia. Blood 1991;78:1096104.
Kelton JG, Bebenek G. Granulocytes do not have sur- 74.
face ABO antigens. Transfusion 1985;25:5679.
Dunstan RA, Simpson MB, Sanflippo FP. Absence of 75.
specifc HLA-DR antigens on human platelets and neu-
trophils (abstract). Blood 1984;64:85a.
Thorsby E. HLA antigens on human granulocytes stud- 76.
ied with cytotoxic iso-antisera obtained by skin graft-
ing. Scan J Haematol 1969;6:11927.
Sachs UJ, Bux J. TRALI after transfusion of cross-match- 77.
positive granulocytes. Transfusion 2003;43:16836.
Stroncek DF, Leonard K, Eiber G, et al. Alloimmuni- 78.
zation after granulocyte transfusions. Transfusion
1996;36:100915.
McCullough J, Clay M, Hurd D, et al. Effect of leuko- 79.
cyte antibodies and HLA matching on the intravascular
recovery, survival, and tissue localization of 111-indium
granulocytes. Blood 1986;67:5228.
M.E. Clay et al.
Byum A. Isolation of mononuclear cells and granu- 80.
locytes from blood. Isolation of mononuclear cells by
one centrifugation and of granulocytes by combining
centrifugation and sedimentation at 1g. Scan J Clin Lab
Invest 1968;21(Suppl 97):7789.
English D, Anderson BR. Single-strep separation of red 81.
blood cells, granulocytes, and mononuclear leukocytes
on discontinuous density gradients of Ficoll-Hypaque.
J Immunol Methods 1975;5:24952.
Ferrante A, Thong YH. A rapid one-step procedure for 82.
purifcation of mononuclear and polymorphonuclear
leukocytes from human blood using a modifcation of
the Hypaque-Ficoll technique. J Immunol Methods
1978;24:38993.
Ferrante A, Thong HY. Optimal conditions for simultaneous 83.
purifcation of mononuclear and polymorphonuclear
leukocytes from human blood by the Hypaque-Ficoll
method. J Immunol Methods 1980;36:10917.
Madyastha P, Madyastha KR, Wade T, et al. An im- 84.
proved method for rapid layering of Ficoll-Hypaque
double density gradients suitable for granulocyte sepa-
ration. J Immunol Methods 1982;42:2816.
Lucas G, Rogers S, de Haas M, et al. Report on the 85.
Fourth International Granulocyte Immunology Work-
shop: progress toward quality assessment. Transfusion
2002;42:4628.
Lalezari P, Pryce S. Detection of neutrophil and plate- 86.
let antibodies in immunologically induced neutrope-
nia and thrombocytopenia. In: Rose NR, Friedman H,
eds. Manual of clinical immunology. Washington, DC:
American Society of Microbiology 1980;7449.
McCullough J, Clay ME, Press C, et al. Granulocyte se- 87.
rology: a clinical and laboratory guide. Chicago: ASCP
Press, 1988:1214, 16871.
Verheugt FWA, Von dern Borne AEGKr, Decary S, et 88.
al. The detection of granulocyte alloantibodies with
an indirect immunofuorescence test. Br J Haematol
1977;36:53344.
Kiefel V, Santoso S, Weisheit M, et al. Monoclonal anti- 89.
body-specifc immobilization of platelet antigens (MAI-
PA): a new tool for the identifcation of platelet reactive
antibodies. Blood 1987;70:17226.
Minchinton R, Noonan K, Johnson TJ. Examining 90.
technical aspects of the monoclonal antibody im-
mobilization of granulocyte antigen assay. Vox Sang
1997;73:8792.
Stroncek D. Neutrophil alloantigens. Transfusion Med 91.
Rev 2002;16:6775.
Wang E, Marincola FM, Stroncek D. Human leukocyte 92.
antigens and human neutrophil antigens systems. In:
Hoffman R, Benz E, Shattil Sj, et al., eds. Hematology:
basic principel and practices 4th ed. Philadelphia: El-
sevier, 2005:240122.
Kissel K, Scheffer S, Kerowgan M, et al. Molecu- 93.
lar basis of NB1 (HNA-2a, CD177) defciency. Blood
2002;99:42313.
Vigrali DA. Multiplexed particle-based fow cytometric 94.
assays. J Immunol Methods 2000;243:24355.
Stroncek DF, Adams S, Lee JH, et al. Neutrophil anti- 95.
body testing. ASHI Q 2008;32:6872.
Bux J, Kissel K, Hofmann C, et al. The use of allele-spe- 96.
cifc recombinant Fc receptor IIIb antigens for the de-
tection of granulocyte antibodies. Blood 1999;93:357
62.
Yasui K, Miyazaki T, Matsuyama N, et al. Establish- 97.
ment of cell lines stably expressing HNA-1a, 1b, and
-2a antigen with low background reactivity in fow cy-
tometric analysis. Transfusion 2007;47:47886.
Yasui K, Hirayama F, Matsuyama N, et al. New cell lines 98.
express HNA-1c, -4a, -4b, -5a, or -5b for identifcation
of HNA antibodies. Transfusion 2008;48:10379.
Lucas GF, Carrington PA. Results of the First Inter- 99.
national Granulocyte Serology Workshop. Vox Sang
1990;59:2516.
Bux J, Chapman J. Report on the Second Interna-
tional Granulocyte Serology Workshop. Transfusion
1997;37:97783.
ISBT Working Party on Granulocyte Immunobiol-
ogy. Recommendations of the ISBT Working Party on
Granulocyte Immunobiology for leukocyte antibody
screening in the investigation and prevention of
antibody-mediated transfusion-related acute lung in-
jury. Vox Sang 2009;96:2667.
Mary E. Clay, MS, MT(ASCP), (corresponding author) Di-
rector, Transfusion Medicine, Research and Development
Program, University of Minnesota Medical School, De-
partment of Laboratory Medicine and Pathology, MMC
198, D-251 Mayo Building, 420 Delaware Street, SE, Min-
neapolis, MN 55455, and Consultant, Neutrophil & Plate-
let Immunology Laboratory, American Red Cross, Mid-
America Blood Services Region, 100 South Roberts Street,
St. Paul, MN 55107, Randy M. Schuller, BS, MT(ASCP),
Supervisor, and Gary J. Bachowski, MD, PhD, Assistant
Medical Director, Neutrophil and Immunobiology Labo-
ratory, American Red Cross, Mid-America Blood Services
Region, St. Paul, MN.
100.
101.

For information concerning the National
Reference Laboratory for Blood Group
Serology, including the American Rare
Donor Program, please contact Sandra
Nance, by phone at (215) 451-4362, by fax at
(215) 451-2538, or by e-mail at snance@usa.
redcross.org
A review of the Colton blood group system
History
The idea that there must be openings in the cell mem-
brane to permit the fow of water and salts to maintain cel-
lular function and to eliminate the byproducts of cellular
metabolism was recognized by the middle of the 19th cen-
tury. Water is the major component of all living cells, in-
cluding those of vertebrates, invertebrates, and unicellular
organisms and plants. Salt water comprises approximately
70 percent of the human body. After debating for decades
about the process of water transfer across cell membranes,
most scientists fnally decided that water must pass free-
ly through biologic membranes by simple diffusion. Only
a small number of scientists disagreed, and they had only
indirect evidence that there were special channels for the
transport of water. After all, there must be something to
explain the high permeability of RBCs and renal tubules.
Moreover, the water permeability of these tissues can be re-
versibly inhibited by mercuric ions.
1
By 1950 it had been shown that water moves quickly
into and out of cells through pores that are selective for
water molecules only. Every second, billions of water mol-
ecules pass through a single channel. By the mid-1980s
scientists had found an RBC membrane protein whose 28-
kDa N-terminal amino acid sequence was related to the
26-kDa major intrinsic protein (MIP26) of bovine lens f-
ber cells.
2,3
The 28-kDa protein was shown to be a unique
molecule that is abundant on erythrocytes and on renal tu-
bules. Preston and Agre
4
succeeded in 1991 in isolating the
same membrane protein from RBCs, which in 1997, was of-
fcially named aquaporin-1 (AQP1) by the Human Genome
Organization, for the water channel. AQP1 has been found
in many tissues including several parts of the kidney, liver,
gall bladder, eye, choroid plexus, hepatobiliary epithelium,
and capillary endothelium. AQP1 turned out to be the frst
of the family of major intrinsic proteins (MIP), which are
highly conserved membrane proteins that regulate small
molecule transport. The MIP family of proteins is believed
to date back 2.5 to 3 billion years in evolutionary time.
5
It is
believed that the MIP family evolved from two basal lineag-
es: AqpZ-like water channels and GlpF-like glycerol facili-
tators. These divergent lineages probably originated from
an archaeal (AqpM-like) aquaporin.
5
The MIP family is now
divided into two groups: aquaporins (AqpZ, AQP0, AQP1,
AQP2, AQP4, AQP5, AQP6, and AQP8) and aquaglycerol-
porins (GlpF, AQP3, AQP7, AQP9, and AQP10).
610
In 1994,
AQP1 protein was characterized as carrying the Colton
blood group antigens.
11
Genetics and Inheritance
In 1965 in Oslo, Norway, the discovery of an antibody
that detected a public (or high prevalence) antigen was
linked to two other cases discovered earlier by workers in
Minneapolis, Oxford, and London. The antigen was named
Co
a
(Colton
a
) in 1967 for the frst of the three producers of
anti-Co
a
, who was from Minneapolis; it should actually have
been named Calton, but the handwriting on the tube ap-
parently was misread.
12,13
Co
a
was shown to be an inherited
characteristic.
14
In 1970, the antithetical antigen, Co
b
, was
reported by Giles et al (Table 1).
15
In analysis of unrelated
people of European extraction, the occurences of the three
major Colton phenotypes were as follows: Co(a+b) 0.914,
Co(a+b+) 0.084, and Co(ab+) 0.002 (Table 2).
16
In 1971,
the genetic independence of the Colton blood group system
from Kell and Lutheran was reported, and it was confrmed
that it is independent from MNSs, P, Rh, Duffy, Dombrock,
and sex.
17
Co
a
and Co
b
are codominant alleles that respec-
tively encode the Co
a
and Co
b
antigens at the RBC surface. In
1974 the Co(ab) phenotype of three persons in a French
Canadian family was reported.
18
The serum from a patient
in 1964 (Swarts) was eventually found to be negative only
with these Co(ab) family members.
16
Colton is the 15th
human blood group system recognized by the ISBT (ISBT
015).
Molecular Basis
The AQP1 locus was mapped to human chromosome
7p14 by fuorescent in situ hybridization.
19
The Colton (Co)
blood group antigens had been previously linked to the
short arm of chromosome 7.
20
Immunoprecipitation studies
Review
Table 1. Antigens of the Colton blood group system: their preva-
lence, molecular basis, and sensitivity to enzyme and DTT treatment
Name
ISBT
symbol
ISBT
number Prevalence
Molecular
basis
Effect of
enzymes
Co
a
CO1 015001 99.8% Ala 45 PR TR CR
NR DTTR
Co
b
CO2 015002 8.6% Val 45 PR TR CR
NR DTTR
Co3 CO3 015003 >99.99% unknown PR TR CR
NR DTTR
CR =
1
-chymotrypsin resistant; DTTR = dithiothreitol resistant;
NR = neuraminidase resistant; PR = papain resistant; TR = trypsin
resistant.
Table 2. Phenotypes (% occurrence)
Phenotype % Occurrence (most
populations)
16
Co(a+b) 91.4
Co(ab+) 0.2
Co(a+b+) 8.4
Co(ab) <0.01
the Co protein have been reported to cause decreased or no
expression of the Colton antigens. An amino acid change
of Gln47Arg was observed to cause a Co(ab+
w
) typing.
29

The same amino acid substitution was recently fully inves-
tigated in a Turkish female with the Co(ab) phenotype.
30

Her RBCs were found by molecular studies to have normal
water permeability by stopped-fow analysis and a normal
amount of AQP1 protein in the RBC membrane. (See sec-
tion on clinical signifcance). By PCR-RFLP the specimen
was genotyped as Co
a
/Co
a
, but full sequencing revealed a
140A>G mutation leading to a Gln47Arg amino acid sub-
stitution. This change in amino acids is close to the Co
a
/Co
b

polymorphism, and thus, the probable cause of the appar-
ent discrepancy between phenotype and genotype. When
using a specifcally designed K-562 cell expression system
for Colton antigens, the Gln47Arg substitution was
shown to inhibit both Co
a
and Co
b
antigen expression.
29

The antibody produced by these Co(ab) individuals
with a Gln47Arg substitution is not a mixture of anti-Co
a

and -Co
b
, rather it is anti-Co3like, reacting with both
Co(a+b) and Co(ab+) RBCs, but more weakly than anti-
Co3 produced by the true null phenotype. This suggests the
existence of a new high-prevalence antigen in the Colton
blood group system, with a corresponding antibody that
Arnaud et al. proposed to name anti-Co4.
30
Therefore, it is
possible that some of the Co(ab) people known through-
out the world, who would have not been fully investigated
by molecular techniques, may actually not be true Co
null

but CO:1,2,3,4. This may also be why some antibod-
ies directed against a high-prevalence Colton antigen, mis-
takenly considered as anti-Co3, were unexpectedly weakly
reactive with Co:3 RBCs.
31
Biochemical Physiology of the Colton Glycoprotein
Protons are necessary for the bioenergetics of the living
cell. The proton gradient drives many transport functions,
membrane fusions, and the synthesis of ATP. Maintaining
the proton gradient is essential, and indiscriminate pro-
ton leakage across cellular membranes would be fatal to
the living cell. At the same time, cells must maintain the
balance of waters ability to permeate the membrane bi-
layer. Aquaporins are strictly selective water chan-
nels that have evolved for that purpose. AQP1 was
found to have a confguration with intracellular N
and C terminals, similar to that of other ion chan-
nel proteins. Protein modeling from the cDNA
sequences predicts a two-tandem repeat of three
membrane-spanning helices that exist as tetram-
ers within the cell membrane (Fig. 2). An electro-
static feld created by the highly conserved amino
acid sequence NPA forms the water channel, which
prevents proton loss while allowing water to perme-
ate the membrane. The most prominent feature is
the association of loop B with loop E driven by their
positive N-terminal ends that brings together the
NPA motifs at the center of the pore. This causes
the formation of strong electrostatic felds that block
Colton blood group system review
and DNA sequencing showed that the Colton antigens are
the result of a polymorphism in AQP1 at the extracellular
site of loop A, which connects the frst and second spanning
domains. Figure 1 depicts the structure of the protein as it
passes through the membrane, and indicated are the Co
a
/
Co
b
polymorphic site at position 45 and the only N-glycan
site at position 42. The Co
a
/Co
b
polymorphism arises from
a single amino acid change of Ala45Val (alanine for Co
a
, va-
line for Co
b
) caused by a nucleotide change of C>T at posi-
tion 134 in exon 1 of AQP1 gene, named CO according to
the ISBT nomenclature and AQP1 according to the Human
Genome Organization (HUGO).
11
The Colton gene consists
of four exons distributed over 11.6 kbp of genomic DNA.
The amino acid sequence of the wild-type AQP1 protein is
shown in Table 3. Also known as the channel-forming inte-
gral protein (CHIP), the Colton glycoprotein is a multipass
protein consisting of three external loops and six mem-
brane-spanning regions, as well as 2 cytoplasmic loops,
with both the NH
2
and COOH terminus located on the cyto-
plasmic side of the membrane.
22
Common to all MIP super-
family proteins is the occurrence of the asparagine-proline-
alanine (NPA) motifs, two being present in each half of the
protein. An hourglass structural model has been proposed
by Agre and colleagues.
23,24
The molecular basis of the null Co(ab) phenotype
has been attributed to several different molecular back-
grounds; these are depicted in Table 4. Other changes in
Out
N-glycosylation site
A = Co
a
V = Co
b
RBC
Membrane
76
78
192
194
C
E
B
D
In
NH
2 COOH
N
P
P
N
A
A
1 2 3 4 5 6
A
42
45
N A/V
Fig. 1 Linear diagram depicting the Colton protein with the six
membrane-spanning domains with both the N terminus and C
terminus inside the cell membrane. The Co
a
/Co
b
polymorphism is
indicated at position 45, as is the glycosylation site at N42. The
signature NPA motifs are depicted on the internal loops B and D.
21
Table 3. Amino acid sequence of wild-type AQP1 protein
13
MASEFKKKLF WRAVVAEFLA TTLFVFISIG SALGFKYPVG NNQTAVQDNV 50
KVSLAFGLSI ATLAQSVGHI SGAHLNPAVT LGLLLSCQIS IFRALMYIIA 100
QCVGAIVATA ILSGITSSLT GNSLGRNDLA DGVNSGQGLG IEIIGTLQLV 150
LCVLATTDRR RRDLGGSAPL AIGLSVALGH LLAIDYTGCG INPARSFGSA 200
VITHNFSNHW IFWVGPFIGG ALAVLIYDFI LAPRSSDLTD RKVWTSGQV 250
EEYDLDADDI NSRVEMKPK 269
Amino acid sequence taken from GenBank, accession # M77829
Also see the Blood Group Antigen Gene Mutation Database (dbRBC)
at http://www.ncbi.nlm.nih.gov/gv/mhc/xslcgi.cgi?cmd=bgmut/home
(accessed April 13, 2010).
the permeation of protons as well as H
3
O+ and other cat-
ions. When Xenopus laevis oocytes were transfected with
cRNA, they were found to have markedly increased water
permeability, causing the oocytes to swell and burst.
32
This
permeability could be reversibly inhibited by Hg2+.
33
When
the same cRNA transcript was reconstituted into lipid vesi-
cles, the same observations were made. This confrmed that
membranes exhibit water permeability and showed that the
water permeability of membranes with AQP1 was up to 100
times greater than that of those without it.
Colton Antigen Typing
To date, anti-Co
a
and anti-Co3 reagents, either mono-
clonal or polyclonal, are not commercialized. As a result,
human polyclonal reagents are only available in immu-
nohematology reference laboratories, but usually in small
amounts. However, a human polyclonal anti-Co
b
is avail-
able in Europe as a CE-marked in vitro diagnostic reagent
(DiaMed, a division of Bio-Rad, Cressier sur Morat,
Switzerland). The two major high-throughput molecu-
lar genotyping platforms, a commercialized DNA glass array
by Progenika (Barcelona, Spain) and a DNA bead array by
BioArray/Immucor (Norcross, GA) allow Co
a
and Co
b
al-
lele typing by searching for the 134C>T polymorphism in
exon 1, with subsequent prediction of Colton phenotype. It
is important, however, to be aware that prediction of the Co
phenotype in Co
null
and Co
variant
people would be a source of
misinterpretation with these DNA-based techniques.
Clinical Signifcance
Antibodies to Colton blood group antigens are gener-
ally IgG and react best by the antiglobulin test, especially
when protease-treated RBCs are used.
34,35
Some anti-Co
a
,
-Co
b
and -Co3 may bind complement.
36
Few reports of sig-
nifcant delayed or acute transfusion reactions or hemolytic
disease of the fetus and newborn (HDFN) attributable to
anti-Co
a
have been reported, although both are known to
occur with severe morbidity.
27,3740

Anti-Co
b
is relatively rare and often found in patients
sera containing other alloantibodies.
34,35
Two reports of in
Table 4. Molecular bases of the Colton-null [Co(ab)] phenotype: effect of the change at the DNA and protein level, and ancestry
of the probands
Molecular alteration Effect
Observed result on
AQP1 protein
Presence
of anti-Co3
HDFN
reported Ancestry Reference
Co
null
phenotypes
Large deletion compris-
ing most of exon 1
Absent or aberrant
transcript
AQP1 not detected in
RBC membrane extracts
Not reported No Northern
European
Proband 1 in Preston
et al.
25
c.309insT Frameshift mutation AQP1 not detected in
Yes No French Proband 2 in Preston
et al.
25
c.113C>T Missense mutation
(p.Pro38Leu)
<1% AQP1 expression in
Not reported No Northern
European
Proband 3 in Preston
et al.
25
c.576C>A Missense mutation
(p.Asn192Lys)
Not tested Yes No Portuguese Chrtien et al.
26
c.232delG Frameshift mutation Not tested Yes Yes Indian Joshi et al.
27
c.601delC Frameshift mutation Not tested Yes No Caucasian Nance et al.
28
; Reid
and Lomas-Francis
13
Altered Co phenotypes
c.140A>G Missense mutation
(p.Gln47Arg)
Not tested No No Not reported,
presumably
Caucasian
Wagner and Flegel
29
c.140A>G Missense mutation
(p.Gln47Arg)
Normal AQP1 expression
in RBC membrane
No No Turkish Arnaud et al.
30
AQP1 = aquaporin 1; HDFN = hemolytic disease of the fetus and newborn.
According to the recommendations of the Human Genome Variation Society and National Institutes of Health Single Nucleotide Poly-
morphism database, c. means that the position of the polymorphism refers to the complementary DNA, p means that the position
of the polymorphism refers to the protein.
Out
RBC
Membrane
In
NH
2
COOH
1 2 3 4 5 6
H
2
O
N
N
A
A
P
P
Fig. 2 The AQP1 water channel is shown as an hourglass confgu-
ration created by the association of the six transmembrane domains
and the crossing of loops B and E with the electrostatic feld pro-
duced by the NPA motifs forming the water channel.
21
vivo survival studies of
51
Cr-labeled Co(b+) RBCs in pa-
tients with anti-Co
b
showed survival at 94 percent and 85
percent, respectively, at 1 hour, decreasing to 51 percent at
24 hours and 10 percent at 96 hours.
41,42
Co
b
is fully devel-
oped at birth, but no signifcant HDFN has been reported to
date.
34
Anti-Co3 was reported to cause severe HDFN requir-
ing neonatal transfusion.
43
Transfusion of incompatible
Co(a+b) RBCs in a patient with anti-Co3 was reported to
be responsible for a mild hemolytic transfusion reaction.
27

There are also reports of autoantibodies mimicking Colton
specifcities.
44
AQP1-defcient individuals were studied for their renal
function and capillary permeability before and after water
deprivation.
45
Baseline studies did not reveal any abnormali-
ties; however, after water deprivation this study revealed an
inability to concentrate urine. Thus, Colton-null individuals
may be subject to serious hydroelectrolytic and metabolic
disorders should they become severely dehydrated.
Acknowledgments
The authors thank Robert Ratner for preparation of the
manuscript and fgures, and Hallie Lee-Stroka for her criti-
cal review and helpful comments. This work was funded in
part by a grant from the MetLife Foundation (GH).
References
Macey RI. Transport of water and urea in red blood 1.
cells. Am J Physiol 1984;246:C195203.
Agre P, Saboori AM, Asimos A, Smith BL. Purifcation 2.
and partial characterization of the Mr 30,000 integral
membrane protein associated with the erythrocyte
Rh(D) antigen. J Biol Chem 1987;262:17497503.
Saboori AM, Smith BL, Agre P. Polymorphism in the 3.
Mr 32,000 Rh protein purifed from Rh(D)-positive
and -negative erythrocytes. Proc Natl Acad Sci U S A
1988;85:40425.
Preston GM, Agre P. Isolation of the cDNA for eryth- 4.
rocyte integral membrane protein of 28 kilodaltons:
member of an ancient channel family. Proc Natl Acad
Sci U S A 1991;88:111104.
Kozono D, Ding X, Iwasaki I, et al. Functional expres- 5.
sion and characterization of an archaeal aquaporin.
AqpM from Methanothermobacter marburgensis. J
Biol Chem 2003;278:1064956.
Roudier N, Ripoche P, Gane P, et al. AQP3 defciency in 6.
humans and the molecular basis of a novel blood group
system, GIL. J Biol Chem 2002;277:458549.
Vajda Z, Pedersen M, Fchtbauer EM, et al. Delayed on- 7.
set of brain edema and mislocalization of aquaporin-4
in dystrophin-null transgenic mice. Proc Natl Acad Sci
U S A 2002;99:131316.
Ikeda M, Beitz E, Kozono D, Guggino WB, Agre P, Yasui 8.
M. Characterization of aquaporin-6 as a nitrate channel
in mammalian cells. Requirement of pore-lining resi-
due threonine 63. J Biol Chem 2002;277:398739.
Carbrey JM, Gorelick-Feldman DA, Kozono D, Praeto- 9.
rius J, Nielsen S, Agre P. Aquaglyceroporin AQP9: sol-
ute permeation and metabolic control of expression in
liver. Proc Natl Acad Sci U S A 2003;100:294550.
Endeward V, Musa-Aziz R, Cooper GJ, et al. Evidence 10.
that aquaporin 1 is a major pathway for CO
2
transport
across the human erythrocyte membrane. FASEB J
2006;20:197481.
Smith BL, Preston GM, Spring FA, Anstee DJ, Agre P. 11.
Human red cell aquaporin CHIP. I. Molecular charac-
terization of ABH and Colton blood group antigens. J
Clin Invest 1994;94:10439.
Heist H, van der Hart M, Madsen G, et al. Three ex- 12.
amples of a new red cell antibody, anti-Co
a
. Vox Sang
1967;12:1824.
Reid ME, Lomas-Francis C. The blood group antigen 13.
factsbook. 2nd ed. San Diego, CA: Academic Press,
2004.
Wray E, Simpson S. A further example of anti-Co 14.
a
and
two informative families with Co(a) members. Vox
Sang 1968;14:1302.
Giles CM, Darnborough J, Aspinall P, Fletton MW. 15.
Identifcation of the frst example of anti-Co
b
. Br J Hae-
matol 1970;19:2679.
Race RR, Sanger R. Blood groups in man. 6th ed. Ox- 16.
ford, England: Blackwell Scientifc Publications, 1975.
Lewis M, Kaita H, Anderson C, Chown B. Independence 17.
of Colton blood group. Transfusion 1971;11:2234.
Rogers MJ, Stiles PA, Wright J. A new minus-minus 18.
phenotype: three Co(ab) individuals in one family
(abstract). Transfusion 1974;14:508.
Moon C, Preston GM, Griffn CA, Jabs EW, Agre P. 19.
The human aquaporin-CHIP gene. Structure, orga-
nization, and chromosomal localization. J Biol Chem
1993;268:157728.
Zelinski T, Kaita H, Gilson T, Coghlan G, Philipps S, 20.
Lewis M. Linkage between the Colton blood group locus
and ASSP11 on chromosome 7. Genomics 1990;6:6235.
Agre P, King L, Yasui M, et.al. Topical Review: Aqua- 21.
porin water channelsfrom atomic structure to clinical
medicine. J Physiol. 2002;542:316.
King LS, Agre P. Pathophysiology of the aquaporin wa- 22.
ter channels. Annu Rev Physiol 1996;58:61948.
Mathai JC, Agre P. Hourglass pore-forming domains 23.
restrict aquaporin-1 tetramer assembly. Biochemistry
1999;38:9238.
Jung JS, Preston GM, Smith BL, Guggino WB, Agre 24.
P. Molecular structure of the water channel through
aquaporin CHIP. The hourglass model. J Biol Chem
1994;269:1464854.
Preston GM, Smith BL, Zeidel ML, Moulds JJ, Agre P. 25.
Mutations in aquaporin-1 in phenotypically normal hu-
mans without functional CHIP water channels. Science
1994;265:15857.
Colton blood group system review
Chrtien S, Catron JP, de Figueiredo M. A single mu- 26.
tation inside the NPA motif of aquaporin-1 found in a
Colton-null phenotype (letter). Blood 1999;93:40213.
Joshi SR, Wagner FF, Vasantha K, Panjwani SR, Fle- 27.
gel WA. An AQP1 null allele in an Indian woman with
Co(ab) phenotype and high-titer anti-Co3 associated
with mild HDN. Transfusion 2001;41:12738.
Nance S, Kavitsky DL, Meny G, Vege, S, Westhoff CM. 28.
An example of anti-Co3 not causing hemolytic disease
of the newborn. Transfusion 2002;42(Suppl):105S.
Wagner FF, Flegel WA. A clinically relevant Co(a)-like 29.
allele encoded by AQP1 (Q47R) (abstract). Transfusion
2002;42(Suppl):24S5S.
Arnaud L, Helias V, Menanteau C, Peyrard T, et al. A 30.
functional AQP1 allele producing a Co(ab) pheno-
type revises and extends the Colton blood group sys-
tem. Transfusion 2010. In press.
Lacey PA, Robinson J, Collins ML, et al. Studies on the 31.
blood of a Co (ab) proposita and her family. Transfu-
sion 1987;27:26871.
Preston GM, Carroll TP, Guggino WB, Agre P. Appear- 32.
ance of water channels in Xenopus oocytes expressing
red cell CHIP28 protein. Science 1992;256:3857.
Kozono D, Yasui M, King LS, Agre P. Aquaporin wa- 33.
ter channels: atomic structure and molecular dynamics
meet clinical medicine. J Clin Invest 2002;109:13959.
Daniels G. Human blood groups. 2nd ed. Malden, MA: 34.
Blackwell Science, 2002.
Issitt PD, Anstee DJ. Applied blood group serology. 4th 35.
ed. Durham, NC: Montgomery Scientifc Publications,
1998.
Covin RB, Evans KS, Olshock R, Thompson HW. Acute 36.
hemolytic transfusion reaction caused by anti-Co
a
. Im-
munohematology 2001;17:459.
Simpson WKH. Anti-Co 37.
a
and severe haemolytic disease
of the newborn. S Afr Med J 1973;47:13024.
Agre P, Smith BL, Baumgarten R, et al. Human red cell 38.
aquaporin CHIP. II. Expression during normal fetal de-
velopment and in a novel form of congenital dyseryth-
ropoietic anemia. J Clin Invest 1994;94:10508.
Kurtz SR, Kuszaj T, Ouellet R, Valeri CR. Survival of ho- 39.
mozygous Co
a
(Colton) red cells in a patient with anti-
Co
a
. Vox Sang 1982;43:2830.
Michalewska B, Wielgos M, Zupanska B, Bartkowiak R. 40.
Anti-Co
a
implicated in severe haemolytic disease of the
foetus and newborn. Transfus Med 2008;18:713.
Dzik WH, Blank J. Accelerated destruction of ra- 41.
diolabeled red cells due to anti-Colton
b
. Transfusion
1986;26:2468.
Hoffmann JJML, Overbeeke MAM. Characteristics of 42.
anti-Co
b
in vitro and in vivo: a case study. Immunohe-
matology 1996;12:1113.
Savona-Ventura C, Grech ES, Zieba A. Anti-Co3 and se- 43.
vere hemolytic disease of the newborn. Obstet Gynecol
1989;73:8702.
Moulds M, Strohm P, McDowell MA, Moulds J. Autoan- 44.
tibody mimicking alloantibody in the Colton blood group
system (abstract). Transfusion 1988;28(Suppl):36S.
King LS, Choi M, Fernandez PC, Cartron JP, Agre P. De- 45.
fective urinary-concentrating ability due to a complete
defciency of aquaporin-1. N Engl J Med 2001;345:1759.
Gregory R. Halverson, BS, MT(ASCP)SBB, DLM (cor-
responding author), Laboratory of Immunochemistry,
New York Blood Center, 310 East 67th Street, New York,
NY 10065, and Thierry Peyrard, PharmD, MS, European
Specialist in Clinical Chemistry and Laboratory Medicine,
National Immunohematology Reference Laboratory, Na-
tional Institute of Blood Transfusion, 20 rue Bouvier 75011
Paris, France.
Free Classifed Ads and Announcements
Immunohematology will publish classifed ads and announcements (SBB schools, meetings, symposia, etc.)
without charge. Deadlines for receipt of these items are as follows:
Deadlines
1st week in January for the March issue
1st week in April for the June issue
1st week in July for the September issue
1st week in October for the December issue
E-mail or fax these items to immuno@usa.redcross.org or (215) 451-2538.
Alloimmunization to the D antigen by a
patient with weak D type 21
Antibodies of apparent D specificity may be found in D+ pa-
tients. We report a D+, multi-transfused Caucasian woman
with myelodysplasia who exhibited several alloantibodies.
One antibody was a moderately strong (2+) anti-D that per-
sisted for 9 months, until the woman died. Molecular analy-
sis of the patients RHD gene identified the rare weak D type
21 (938C>T) allele. D alloantibodies do not occur in patients
with most weak D types, but some patients with a weak D
phenotype, including those with type 21, can produce an-
tibodies to nonself epitopes of the wild-type D antigen.
Key Words: alloimmunization, anti-D, RHD gene, weak D
A
ntibodies with D specifcity are unusual in D+ pa-
tients. Alloantibodies to D may be (1) passively
transfused in donor plasma products, (2) passively
transferred but actively produced by lymphocytes carried
in transplanted organs, and (3) actively produced by an al-
loimmunized patient. Autoantibodies to D may be warm- or
cold-reactive antibodies with a preference for D+ RBCs or
may be antibodies to LW that mimic anti-D specifcity.
1
In practice, the combination of clinical, serologic, and
molecular investigations helps to determine why a D+ pa-
tient has anti-D. Establishing the source of a D+ patients
alloantibody or autoantibody with D specifcity is important
in deciding whether D+ or D units should be transfused to
optimize donor RBC survival (Table 1).
In this case report, we describe a woman whose RBCs
were strongly D+ at the phenotypic level, but had an RHD
genotype consistent with weak D, and an alloanti-D.
Case Report
A 72-year-old, gravida 3, Caucasian woman with myel-
odysplasia and pancytopenia was transferred from another
hospital, where in 6 months of treatment, she had received
an unknown number of platelet transfusions, but no RBC
transfusions.
In our hospital and before any RBC transfusion, the
patients RBCs serologically typed as D+ (4+) by tube ag-
glutination at immediate spin phase, using a monoclonal-
polyclonal blended anti-D (Ortho Clinical Diagnostics Co.,
Raritan, NJ). During a disease course of 13 months, the
patient received 37 apheresis-processed platelet units and
108 leukocyte-reduced units of RBCs, which produced al-
loimmunization to K, E, and C
w
. After 9 months, her serum
also demonstrated a 2+ anti-D in a gel-based, antibody de-
tection system (Micro Typing Systems, Inc., Ortho Clinical
Diagnostics, Pompano Beach, FL). The initial DAT (tube,
anti-IgG) and all but one subsequent DAT were negative as
was the auto-control. On the one occasion, the DAT became
weakly positive in a mixed-feld reaction. Acid eluates pre-
pared from the patients RBCs contained only anti-E. There
was no history of organ transplantation or administration
of a plasma product that might contain anti-D. The anti-D
persisted throughout the patients course. Blood samples
were collected and referred for RHD and RHCE gene analyses.
Although continued transfusions with D blood were
prescribed, the patient refused all further medical treat-
ment. Her myelodysplasia progressed to early acute leuke-
mia before she expired at home.
Molecular Test Results
Molecular tests were performed at the Molecular Red
Cell and Platelet Testing Laboratory of the American Red
Cross, Penn-Jersey Region, and the University of Penn-
sylvania DNA Sequencing Facility, both in Philadelphia,
Case Report
Table 1. Evaluation and transfusion of patients with D+ RBCs and
serum anti-D
Etiology/
pathogenesis Diagnostic evidence
Preferred RBC
transfusion
Passive alloantibody History of transfusion
with D RBC, plasma,
IVIG, RhIG, other product
with anti-D
D until anti-D
wanes
Alloantibodies from
passenger lymphocytes
in donor organ
Organ transplant from D
donor to D+ recipient
D until anti-D
wanes
Autoantibody to D
antigen
Positive direct/indirect
antiglobulin test
(anti-IgG)
D over D+
Anti-D specifcity in
eluate or serum
History: pregnancy, warm
autoimmune hemolytic
anemia, malignancy
Autoantibodies or
alloantibodies to LW
Antibody fails to react
with RBC treated with
dithiothreitol
D over D+
Antibody is absorbed by
D RBC
Alloantibody to epitope
absent from normal D
protein
Anti-D fails to react with
RBC of person who
produced it
D only
Anti-D fails to react with
D RBC
Pennsylvania. The patient was found to be heterozygous
for RHD. No changes associated with any of the known
partial D phenotypes were identifed. A change (938C>T)
was identifed that is associated with weak D type 21 that is
predicted to encode an intracellular amino acid change of
Pro313Leu. The patients RH genotype could be interpreted
to be RHD*weak D type 21, RHCE*Ce/RHCE*ce.
Discussion
Initially, the fnding of anti-D in a D+ patient was puz-
zling and the several causes listed in Table 1 were consid-
ered. Patients whose anti-D was passively transferred in
plasma products (IVIG, RhIG, etc.) or whose serum anti-
bodies were produced by a donors passenger lymphocytes
may require D RBCs until the acquired anti-D has weak-
ened, usually in a matter of weeks to months.
Passive anti-D, however, was ruled out by history. Be-
fore hospital admission at our institution, the patient had
been transfused only with apheresis platelets, but she had
not been transfused with plasma or cellular products from
D donors, nor had she been infused with IVIG or injected
with RhIG. Persistence of the anti-D was further evidence
that the anti-D did not involve passive immunization.
The transfer of passenger lymphocytes could be ex-
cluded by history as well. The patient had never received a
transplant. Passenger lymphocytes are alloantibody-secreting
B cells that are carried in a transplanted donor organ to a
recipient. An anti-D produced by passenger lymphocytes
arises in the donor by alloimmunization to the D antigen
before transplantation. Donor lymphocytes continue to
manufacture detectable anti-D for about 3 months after
transplantation.
Many patients with autoimmune hemolytic anemia and
mimicking antibodies to Rh antigens are treated medically
and transfusion is avoided altogether. In other patients,
transfusion can be delayed until the concentration of an-
tibodies is reduced by medical treatment. Many mimicking
warm autoantibodies have apparent Rh specifcity, especially
anti-e. Only rare D+ patients will possess warm autoan-
tibodies that mimic anti-D. If a patient with autoantibodies
requires transfusion, antigen-negative donor RBCs may be
preferable because they may survive longer than antigen-
positive RBCs.
2
Antibodies to LW are almost always autoantibodies;
those that are less potent appear to have anti-D specifc-
ity because the number of D antigen sites per RBC corre-
lates with the number of LW antigen sites.
3
Thus, D RBCs,
which have fewer LW antigen sites than D+ RBCs, may not
be agglutinated by an anti-LW. However, anti-LW may be
distinguished from anti-D by its failure to agglutinate RBCs
treated with DTT. If transfusion is required for a patient
with autoanti-LW, D RBCs are preferable to D+ RBCs pro-
viding such a patient is c+.
Patient autoantibodies, including anti-LW, are often
transient, but our patients anti-D persisted for 4 months
at constant serologic strength (2+ by the IgG gel test).
Alloantibodies to Rh antigens are often durable (85%).
Negative DATs and absence of RBC agglutination in auto-
control tests also indicated the presence of D alloantibody.
LW alloantibodies, which can mimic anti-D, arise
only in very rare individuals whose RBCs are LW(ab)
or LW(ab+). Neither autoantibodies nor alloantibodies
to LW were specifcally evaluated in the patient described
here.
Alloantibodies to the D antigen occur in people who are
known to lack one or more epitopes of D (partial D pheno-
type). On the other hand, anti-D alloimmune responses in
patients with other weak D types are unusual because their
RBCs do not appear to lack D epitopes when tested with a
standard battery of anti-D monoclonal reagents.
4
Suspicion that the patient had been alloimmunized to
D antigen was raised by the serologic fnding of negative
DATs and the antibodys persistence. In addition, people
who carry variants of the D antigen that are detectable only
by molecular means are capable of producing antibodies
to wild-type protein. In the present case, fnding a variant
RHD gene supported the clinical and serologic evidence of
D alloimmunization.
Historically, some D+ individuals who made anti-D were
initially considered to express a weak D phenotype (frst
termed D
U
) when tested with human polyclonal reagents.
Serologic evidence of alloimmunization to D in a D+ person
included a negative DAT, a serum antibody, and, for some
partial D phenotypes, the patients RBCs also expressed a
low-prevalence Rh antigen.
5
A D+ patient with alloanti-D
was thought to lack part(s) of the D antigen, i.e., a partial
D type. A person with a partial D phenotype would only
produce antibody to those epitopes of the D antigen missing
from his or her own RBCs. Investigative use of monoclonal
antibody revealed many epitopes of the D antigen and sup-
ported the concept that partial D types lacked one or more
epitopes of their D antigen. The monoclonal antibodies also
extended the identifcation and classifcation of the various
partial D types initially defned by polyclonal anti-D.
Many individuals, however, were found to have RBCs
that agglutinated poorly with common anti-D reagents, but
demonstrated the presence of all known epitopes. These D+
RBCs were called weak D in preference to D
U
. Weak D
RBCs expressed fewer D molecules per RBC, a fnding that
accounted for their weak agglutination with reagent anti-
bodies. Theoretically, individuals who carried these weak
D types should not be able to develop alloantibodies to D.
However, it is virtually impossible to distinguish a person
with a weak D phenotype from a person with a weak par-
tial D phenotype by serologic tests; only the production
of alloanti-D will reveal someone to have a weak partial D
phenotype.
When molecular analysis became available, the RHD
(and RHCE) gene(s) revealed far more extensive and com-
plex polymorphisms than were evident by serologic meth-
ods. Weakly agglutinating D serotypes could be related to
point mutations, nonsense mutations, deletions, and tandem
Anti-D in weak D type 21
rearrangements in the RHD and RHCE DNA. The cause of
weak RBC agglutination was confrmed to be an absence of
epitopes in partial D types, a decreased number of protein
molecules in weak D types, or both. In addition, partial D
phenotypes are expressions of genotypes that encode for
changes in the amino acids on the outer surface of the RBC
where the D protein loops outside the RBC membrane.
Patients whose RBCs express partial D types are often im-
munized to D+ RBCs by transfusion or pregnancy. On the
other hand, genotypes that encode weak D phenotypes are
predicted to result in intracellular or intramembranous
changes in amino acids that, in theory, express all epitopes
of D so that patients whose RBCs express weak D types
are not expected to produce anti-D after immunization by
transfusion or pregnancy. Theory is partly supported by
evidencethe most common weak D variants (types 13),
which constitute 70 to 90 percent of all weak D variants, do
not produce alloanti-D.
6

Molecular examination of the various partial and weak
D types indicated that many serologic phenotypes had more
than one basis, but did not suggest that new epitopes could
be identifed.
7
However, serologic evidence suggested oth-
erwise. Anti-D has been found in some weak D patients
whose D protein is predicted by their DNA sequences to
have altered intramembranous and intracellular amino ac-
ids of the D protein. Thus, some weak D molecular alleles
contain missense mutations that not only reduce the num-
ber of antigens per RBC and cause weak RBC agglutination
by reagent antibodies but also cause a loss of D epitopes
and produce changes in the three-dimensional antigenic
structure of the D protein.
The case presented here indicates that the rare weak
D type 21 is an allelic variant whose patient-carriers are
susceptible to transfusion alloimmunization by wild-type D
antigen. Weak D type 21 is the result of a point mutation
that alters an amino acid thought to be in an intracellular
region of the D protein.
8
The rare allele was frst discovered
in one individual during investigation of 270 known weak
D samples in people from northern Germany,
9
and there
has been one subsequent independent observation of it in
Austria.
10
In both cases, the allele has been found on the
same chromosome (cis) as a RHCE*Ce allele in the haplo-
type RHD
weak D type 2-
RHCE*Ce. The Pro313Leu change in the
translated D protein is in its tenth intracellular helix, and all
37 tested epitopes of the D antigen appear to be present.
9
As
expected, the number of D antigen sites per RBC is reduced
(5.2 10
3
) when compared with the normal number (10
4

to 2.5 10
4
) of sites.
11
This antigen density is the highest
among all known weak D alleles. The high density probably
accounts for the patients 4+ serotype and explains why she
was not identifed as a weak D carrier.
Summary
We evaluated a patient with a strongly expressed D an-
tigen with a posttransfusion antibody to the D antigen. The
antibody had clinical and serologic characteristics of an
alloantibody. After RHD and RHCE genotyping, the patient
was found to have the RHD
weakD type 21
RHCE*Ce/RHCE*ce
genotype (R
1
weak D type 21
r phenotype). Although most patients
with a weak D phenotype do not produce alloanti-D, pa-
tients who carry the rare weak D type 21 in the absence of a
normal RHD allele appear susceptible to the production of
D alloantibodies.
Acknowledgments
The authors thank Connie Westhoff, PhD, and Sunitha
Vege for the molecular analyses.
References
Lu RP, Clark P, Mintz PD. Apparent anti-D in a D+ in- 1.
dividual: a clinical and serologic approach. Transfusion
2007;47:1755.
Leger, RM. Chapter 17: The positive direct antiglobu- 2.
lin test and immune-mediated hemolysis. In: Technical
manual. 16th ed. Roback J, Combs MR, Grossman B,
Hillyer, C, eds. Bethesda, MD: American Association of
Blood Banks, 2008:499524.
Gibbs MB. The quantitative relationship of the Rh-like 3.
(LW) and D antigens of human erythrocytes (letter).
Nature 1966;210:6423.
Reid ME. The Rh antigen D: a review for clinicians. 4.
Blood Bull 2008;10:12.
Westhoff CM. Structure and function of the Rh blood 5.
group antigens. Semin Hematol 2007;44:4250.
Flegel W. The genetics of the Rhesus blood group sys- 6.
tem. Dtsch Artzebl 2007;104(10):A6517.
Wagner FF, Frohmajer A, Ladewig B, et al. Weak D al- 7.
leles express distinct phenotypes. Blood 2000;95:2699
708.
Wagner F. Weak D types. Available at: http://www.un- 8.
iulm.de/~fwagner/RH/RB/weakD.htm. Accessed Feb.
25, 2009.
Le Van Kim C, Mouro I, Cherif-Zahar B, et al. Mo- 9.
lecular cloning and primary structure of the human
blood group RhD polypeptide. Proc Nat Acad Sci U S A
1992;89:109259.
Muller TH, Wagner FF, Trockenbacher A, et al. PCR 10.
screening for common weak D types shows different
distributions in three central European populations.
Polin H, Danzer M, Hofer K, Gassner W, Gabriel C. Ef- 11.
fective molecular RHD typing strategy for blood dona-
tions. Transfusion 2007;47: 13505.
Wagner FF, Gassner C, Muller TH, Schonitzer D, 12.
Schunter F, Flegel WA. Molecular basis of weak D phe-
notypes. Blood 1999;93:38593.
Heather McGann, MT(ASCP) SBB (corresponding author),
Supervisor, Blood Bank, Department of Pathology, and
Robert E. Wenk, MD, Consulting Pathologist, Department
of Pathology, Sinai Hospital, 2401 West Belvedere Avenue,
Baltimore, MD 21215.
A review of the Chido/Rodgers blood group
R. Mougey
. . . and the blind, wise blood bankers said, These an-
tibodies are high-titered and have low avidity, they are red
cell, white cell antibodies, they are clinically insignifcant,
are not reactive with enzyme-treated red cells, and can be
absorbed with white cells! Then the chorus of other blind,
wise blood bankers said, NO, THEY ARE NOT!
BASED ON MEMORIES OF DISCUSSIONS ABOUT THESE ANTIBODIES
BACK IN THE 1970S AND THE FABLE OF THE ELEPHANT AND BLIND WISE
MENR. MOUGEY
It was a delight, an honor, and also a little intimidating
to be asked to write a review of the Chido/Rodgers blood
group system, especially considering that the last author to
pen a review for Immunohematology on this subject was
Carolyn M. Giles, the author of so many important publi-
cations on Chido and Rodgers. Moreover, the recent dis-
coveries about Chido and Rodgers have moved well beyond
the feld of blood group serology to an understanding of the
functional protein and genetics at a molecular level.
My delight was in the opportunity to revisit my early
days in performing antibody identifcation and think again
about the challenges that blood bankers faced then in work-
ing with a group of antibodies that no one could agree about.
Chido and Rodgers were usually lumped in with this group,
which in light of current knowledge is quite understandable.
As I recall even the pronunciation for Chido was a matter of
some dispute, with adherents claiming that Cheedo with
a short i vowel, not Chido with a long i sound, was correct.
It occurred to me that this review should cover a historical
perspective, which might be helpful to those blood bankers
whose testing experience is derived from the newer meth-
ods of gel column or solid-phase technology and who might
be wondering what all the fuss was about. The review will
then cover the current understanding of the serology, func-
tion, and biochemistry, and then the genetics of C4 and the
Chido/Rodgers blood group system. The nomenclature for
the complement system is very confusing even for blood
bankers who deal routinely with a nomenclature for a blood
group that is just as confusing.
Historical Perspective
In the beginning there was confusion and chaos. Dur-
ing the 1960s and 70s, the process for antibody identifca-
tion was usually a straightforward and consistent exercise
with good consensus between individual technical staff and
among laboratories, except there was this group of anti-
bodies that would not cooperate. In the reference labo-
ratory where I worked, we called them the grubbies, but
we were too frustrated by working with them to go much
further than that. The usual experience was to spend an
inordinate amount of time performing every possible test,
which led only to excluding what the antibody could not
be. Then when additional samples were obtained to con-
tinue the investigation, it was not uncommon to fnd that
tests that had appeared to be nonreactive were now reactive
with the new antibody sample or with a freshly collected
sample of the test cell. It took me a while to recognize that
with these antibodies, if you wanted to test for specifcity
by typing the patients RBCs or testing known antigen posi-
tive or negative RBCs, it was best to use the freshest RBCs.
If you wanted to fnd compatible blood, older donor units
were more likely to be nonreactive. That is, it was gener-
ally agreed that there were some patients whose antibod-
ies could not be identifed as any of the then known anti-
bodies and whose reactivity was so weak there was a lot of
disagreement as to which test RBCs were truly nonreactive.
This made it very diffcult to defne and describe the true
characteristics of the antigens and the antibodies and show
that the antigens were expressed on the RBC as an inherited
trait. This variability was partly a result of testing methods,
which relied heavily on a multiphasic set of incubation tem-
peratures and the use of a polyclonal antiglobulin reagent.
In that period, blood bankers went to great lengths to de-
tect and identify room temperaturereactive antibodies as
well as antibodies active at 37C. The use of fresh serum for
testing was usually considered essential, with the end result
that blood bankers were pretty good at detecting clinically
signifcant antibodies but even better at fnding clinically
insignifcant antibodies. Blood bankers got lots of prac-
tice at identifying these clinically insignifcant antibodies,
and neutralizing them with human breast milk, avian egg
albumin, hydatid cyst fuid, saliva, and human and guinea
pig urine. The grubbies defed these efforts, and there was
much discussion, some of it acrimonious, on how patients
with these antibodies ought to be managed. The term high
titer, low avidity was preferred by many if only for its ac-
ronym HTLA, even though everyone knew that these anti-
bodies were not always of high titer or of low avidity. When
such weak antibodies were shown to be of high titer, it de-
fed logic that the antibody should show such poor avidity.
Obviously, the fault could lie with the antigen, either its ori-
entation or its number of sites, or some other property not
yet realized. Again, there was a contradiction in the patient
making a high-titered antibody if there was so little antigen
to stimulate the production of antibodies. The frst glimmer
of understanding in these contradictory fndings was the re-
ports on anti-Chido.
Review
Discovery and Serology of Chido/Rodgers
The Chido antibody was frst studied in a patient in
1962, and in a collaborative effort between laboratories in
Minneapolis, Minnesota, and London, United Kingdom, it
was found to be reactive with nearly all cells tested but was
consistently nonreactive with the RBCs of a London panel
donor. Six additional examples were found, and the anti-
body was named Chido after one of the patients and the
combined results were published by Harris et al. in 1967.
1

Their title is as good a description of the grubbies as any, A
nebulous antibody responsible for crossmatching diffcul-
ties. The authors reported that the antibodies could not be
inhibited by saliva, urine, or hydatid cyst fuid.
During this time, there was much speculation about
these grubby antibodies being RBC-WBC antibodies, that
came from some successful experiments using WBCs to ad-
sorb these antibodies.
2
This whole issue of mistaken identity
was very understandable because of the not uncommon an-
tibody problem known as anti-Bgs. Bg antibodies had been
shown to be hemagglutinins with anti-HLA specifcity, and
the Bg antigens were thought to be remnants of those HLA
antigens expressed on the RBC before the loss of its nucle-
us. Because Bgs were antigens of lower frequency, it would
have been so tidy to have them be the antithetical allele to
Chido as the variable strength of reactivity seemed so simi-
lar. Eventually it was realized that the grubby antibodies
had nothing to do with WBC antigens. There may have been
suffcient residual plasma in the WBC preparations used to
adsorb the antibodies, now known to be capable of inhib-
iting anti-Chido, to lead to this mistaken conclusion. The
speculation that the Chido antigen was a WBC antigen did
lead to studies for possible linkage with HLA, an important
advance as linkage studies had shown no association with
any other blood groups.
Many additional examples of anti-Chido were found,
and it was thought that these antibodies did not seem to
cause obvious destruction of Chido-positive donor RBCs.
The work of Middleton and Crookston
3
published in 1972
provided the clue that led to recognizing the true nature of
the blood group. In an attempt to enhance the reactivity
of the anti-Chido they were studying, inert human plasma
was added to the tests; a common practice of the time was
to add a source of protein to enhance antibody binding for
antihuman globulin (AHG) reactions when using polyspe-
cifc AHG. The authors recognized that when the antibody
reactivity disappeared after the addition of plasma the
Chido antigen must also be in plasma. They then found
that performing Chido antigen typings using the plasma of
the person being typed in a neutralization test was a much
more reliable way to type individuals for the Chido antigen
than testing their RBCs. They also noted that weak Chido
antigen reactivity on an individuals RBCs did not appear
to correlate with the amount of antigen in that persons
plasma. The ability to neutralize anti-Chido was also a use-
ful tool for antibody identifcation, especially in sera with
multiple antibodies. This was the frst way in which Chido
was distinguished from the other so-called HTLA specifci-
ties Knops, Cs
a
, and Yk
a
, although some workers continued
to think of anti-Yk
a
as also showing some inhibition with
plasma.
4
At this point, the Chido antigen was thought to be a
plasma antigen that was adsorbed onto the RBC much like
the Lewis antigens. Once accurate, reproducible antigen
typings were possible, family studies on the inheritance of
the Chido gene led to the discovery that Chido was linked
to HLA.
5
This potential relationship to HLA prolonged the
use of the term RBC-WBC antibodies, but the recognition of
HLA linkage to Chido led to the studies to see whether other
genes known to be linked to HLA would also show linkage
to Chido.
The description of anti-Rodgers, a second antibody re-
active with an RBC antigen of high prevalence that could be
neutralized by plasma, followed in 1976.
6
The antibody was
formed by a transfused patient whose RBCs were Chido-
positive who also made an anti-E. The anti-Rg
a
(as anti-Rg
was referred to at that time) gave weak to 4+ reactions with
E RBCs (not your typical low-avidity antibody). Using
plasma neutralization tests, Longster and Giles found that
3.1 percent of donors could not inhibit the antibody, i.e.,
were Rodgers-negative, although they found some donors
who were only partial inhibitors. Nordhagen et al.
7
were
also able to demonstrate partial inhibition of anti-Chido by
the plasma of Chido-positive subjects, and Giles
8
showed
that partial inhibition was an inherited trait. These fndings
would explain conficting results in tests on Chido/Rodgers
antibodies among laboratories if a single source of plasma
was used in inhibition tests. The common practice now is to
use a pool of inert plasmas when attempting to determine
whether an unknown antibody is neutralizable by human
plasma.
Rodgers was also shown to be linked to HLA in two
studies in 1976 that set the stage for the discovery of the
relationship of Chido and Rodgers to the C4 protein.
9,10
In
1978, ONeill et al.
11
showed that Chido and Rodgers are an-
tigenic determinants expressed on the C4 complement pro-
tein. Then Tilley et al.
12
showed that the Chido and Rodgers
determinants are on the C4d fragment of the C4 molecule.
The small amount of C4d normally found on the RBC mem-
brane is the limiting factor that explains the low avidity of
anti-Chido or anti-Rodgers. In fact, if large amounts of C4d
are bound to the RBC under low ionic conditions, anti-Ch
or anti-Rg can be turned into a strongly reactive, direct ag-
glutinin.
13
The puzzle of incomplete or partial inhibition of
the antibodies is caused by the extreme polymorphism of
the C4 genes with multiple alleles that cause variations in
the Ch or Rg determinants.
All this progress made Chido and Rodgers antibodies
easier to recognize with the exception of Chido-positive
variant individuals who have made anti-Chido to the part
of the Chido antigen that they lack.
14
The further serologic
complexities of Chido/Rodgers specifcity will be discussed
in the section on genetics.
Chido/Rodgers review
The inexorable unraveling of the puzzle of the true na-
ture of the blood group Chido/Rodgers is an excellent ex-
ample of the importance of recognizing anomalous fndings
in the study of antibodies that most blood bankers wished
could just be ignored! The fnal refutation of the term HTLA
came from the revelation that the Knops, McCoy, Swain-
Langley, and York blood group sera recognize antigens that
are carried on the RBC complement receptor CR1.
15,16
How-
ever, this fnding seemed to have brought a level of kinship
to these antibodies with Chido and Rodgers that must have
been satisfying to those who wanted to believe that Chido,
Knops, etc. must be related. Again the low avidity of most
Knops antibodies is probably related to the low copy num-
ber of the individuals CR1 receptor. It also seems likely that
these blood groups will be forever linked in the literature as
those HTLAs when they are studied by aspiring blood bankers.
Clinical Signifcance of Chido/Rodgers
It is well documented that many patients with Chido/
Rodgers antibodies have been safely transfused with ran-
dom RBC units, but there are also reports of signifcant re-
actions to plasma or platelet transfusions in patients with
anti-Chido or -Rodgers,
17,18
a cautionary incident remind-
ing blood bankers not to ignore the potential signifcance of
an antibody based on specifcity alone. Strupp et al.
19
also
pointed out the danger in assuming that antibodies that
appear to be of low avidity may be safely ignored without
further investigation. They reported 4 patients with sickle
cell disease for whom the provision of compatible blood and
beneft of transfusion were delayed because Dombrock an-
tibodies were not initially recognized and identifed. How-
ever, there is a need to determine the best way to detect and
identify these antibodies without incurring unnecessary
delays and expense to the patient, but how much testing is
enough? It sometimes seems that when we had the time,
resources, and people to detect and identify these antibod-
ies, we did not have a clue what they were. Now we know all
about them and do not have the time, people, or resources
to do much with them. In this area as with so many other
blood group issues, perhaps molecular methods will identify
good patient practices, and with the use of microarray tech-
nology, defnitive answers to some of these thorny issues
of cost versus beneft will be found. Until that time, when
patients with known anti-Chido or anti-Rodgers require
RBC transfusions, antibody detection, identifcation, and
crossmatching methods should include plasma-neutraliza-
tion studies. After all, these patients are known antibody
formers. If transfusion of plasma components is required,
there should be some consideration given to the possible
effects on the patient of receiving a large volume of plasma,
such as the formation of precipitating immune complexes
or anaphalaxis.
17,18
Finally, the presence of antibodies to C4
proteins may indicate a potential disease association such
as is seen with the patients who have anti-Ch and anti-Rg
and are C4-defcient.
The story of Chido and Rodgers blood groups has moved
well beyond the crossmatching diffculties, and there is
now a wealth of new material that shows the C4 genes and
proteins to be a fascinating model for the evolution of the
genes for critical biologic processes that can both protect
and attack their host organism.
Structure and Function of the C4 Protein
The C4 protein expressing the Chido or Rodgers
epitopes occurs in two isoforms, called C4A for acidic (for-
merly C4F) and C4B for basic (formerly C4S).
20
Awdeh and
Alper
20
proposed the change from F for fast and S for slow
to conform with the then new International System for Hu-
man Gene nomenclature and because in their experiments
sialic acid was removed from plasma proteins before elec-
trophoresis, resulting in distinct but different bands from
those in the experiments of ONeill et al.,
11
which were the
frst to link the Chido/Rodgers antigen to C4 proteins. C4A
proteins usually carry the Rodgers antigens and C4B, the
Chido antigens, although a number of haplotypes have been
discovered in which these associations were switched.
21,22
C4A or C4B protein is expressed as a single-chain pre-
cursor of 1744 amino acid residues, and the C4A or C4B
amino acid sequences are nearly identical (99%) except near
residues 1000 to 1200.
23
The amino acid substitutions that
occur at residues 1101 to 1106 distinguish C4A from C4B.
Single-nucleotide polymorphisms (SNP) are also found in
the region that distinguishes the alleles of C4A from each
other and from C4B and its alleles. The Chido and Rodgers
antigens are thought to differ by both amino acid substitu-
tion as well as conformational binding sites.
24
After translation, the pro-C4 protein is cleaved into al-
pha, beta, and gamma polypeptide chains that are linked
by disulfde bonds.
25
However, incomplete processing ac-
tually leads to multiple structural forms of the peptide be-
ing secreted in the plasma.
26
The activated complement
component C1 initiates cleavage of 77 amino acids from
the N-terminus of the alpha chain, releasing C4a, and the
remainder of the molecule C4b changes shape, exposing
a thioester bond between the sulfhydryl group of Cys-991
and the carbonyl group of Gln-994. This binding site that
is exposed on activation will allow the formation of either
a covalent ester or an amide linkage.
27,28
Activated C4B is
more effcient than C4A at pathogen cell lysis because of its
histidine residue at position 1106 and its serine residue at
position 1102.
29
The liver is the primary site of C4 production, with
smaller amounts produced by macrophages in other tis-
sues.
30
There is a wide variation in the amount of C4 pro-
duced among individuals. The concentration of C4 in the
plasma varies from 80 to 1000 g/mL with variable con-
centrations of C4 isotypes.
31
Moulds et al.
32
found variation
in patterns of C4 protein levels in individuals that showed
either a complete absence of C4A or C4B, more C4A than
C4B, more C4B than C4A, or equal amounts of both. There
R. Mougey
is some correlation between C4 gene copy number varia-
tions and certain polymorphisms with the level of C4 pro-
tein produced, but the effect at the haplotype level can be
lost in the diploid gene expression.
33
C4 can be activated via the classical pathway or the
mannan-binding lectin pathway, although this may be a
simplistic view as a recent paper demonstrated the poten-
tial for localized release of anaphylatoxins C3a and C4a by
tryptase under certain conditions such as those found in the
airways of asthmatics.
34
Although it is simpler to consider
organized pathways in which complement proteins perform
their complex dance with the other parts of the innate and
humoral immune response, it is likely that the protection
or attack effect is far more chaotic and layered in multiple
mechanisms that activate, block, or break down these pro-
teins (see discussion on disease association).
In the classical pathway, C4 is processed by activated
C1, which removes C4a anaphylatoxin from the alpha chain.
C4a is a mediator of local infammation and can induce the
contraction of smooth muscle, increase vascular permeabil-
ity, and cause histamine release from mast and basophilic
WBCs. C4a levels along with the anaphylatoxins C3a and
C5a in plasma were shown to be elevated during normal
pregnancy as an indication of the increased activation of the
complement system, possibly as a protection against infec-
tion.
35
The remaining C4b fragment forms the enzymes C3
convertase (C4bC2a) and C5 convertase (C3bC4bC2a).
These enzymes catalyze the membrane attack proteins,
which open a channel into the cell membrane. C4 binding
protein (C4bp) is another plasma protein that regulates C4
activity by blocking the formation of and promoting the
decay of the classical pathway C3 convertase. Defciency of
C4bp would be expected to result in increased activation of
C3 and complement activity in response to classical path-
way or lectin pathway activation by immune complex for-
mation, bacterial infections, apoptosis, and other triggering
mechanisms.
36
Activated C4B (C4b-B) binds to its target rapidly but
has a short half-life of less than 1 second and has greater
hemolytic activity than C4A.
28
Activated C4A (C4b-A) is
10 times slower than C4B, but it forms amide bonds with
amino group targets of immune complexes (IC) or protein
antigens. C4B more effciently forms ester bonds with car-
bohydrate antigens, resulting in better clearance of patho-
gens.
Besides activation of the complement cascade via the
classical or lectin pathways, the covalent binding of C4 to
immunoglobulins and to IC enhances the solubilization of
immune aggregates and the clearance of IC by binding to
complement receptor on RBCs (CR-1). This IC binding on
the RBC then acts as a shuttle and delivers the IC for phago-
cytosis by macrophages. Thus C4 plays multiple roles for
both the innate and adaptive immune systems in vertebrate
animals.
37
Part of this multiple role is the normal constant low
level of activation of C3 via the classical pathway and the
alternate pathway. In the classical pathway, it was thought
that specifc antigen binding by an antibody capable of ac-
tivating C1 would initiate the pathway, but Manderson et
al.
38
have shown that C3 is also being activated by C1C4 in
the absence of antibody-antigen complexes. They propose
that although circulating constant low levels of IC could
account for their fndings, the continual circulation of the
normal levels of IgG antibodies may form spontaneous un-
stable Ig aggregates that would provide for a C3 tickover as
well. In their experiments, vascular stasis would encourage
or initiate complement activation and would therefore be
augmented at sites of infammation. This would facilitate
pathogen recognition by allowing the deposition of activated
C3 fragments onto foreign surfaces (host cell surfaces with
complement regulatory proteins would be unaffected).
Thus the various polymorphisms of C4 affect the normal
functioning of C4 and can be affected by the individuals
ability to produce suitable levels of C4 proteins. Defciencies
of C4 proteins whether acquired or inherited can increase
the risk of severe viral and bacterial infections.
33
How high
levels of C4 may also contribute to the risk of diseases is
just beginning to be understood (disease association will be
discussed later).
Genetics of C4 and Chido/Rodgers
The original notation of Ch
a
and Rg
a
implied the ex-
pectation that there would be additional antithetical al-
leles Ch
b
and Rg
b
.
1,6
However, it was soon realized that this
blood group did not ft this pattern of inheritance. Chido,
or Ch, and Rodgers, or Rg, are used to describe the cluster
of determinants that are expressed on the C4A or C4B pro-
teins that can neutralize the antibodies formed by Chido- or
Rodgers-negative individuals. Also individuals can type as
positive with anti-Chido, but form anti-Chido to the Chido
epitopes they lack. Currently nine antigens have been de-
scribed: six of high prevalence for Chido, two of high preva-
lence for Rodgers, and one of low prevalence, WH.
24
Eight
phenotypes for the CH/RG system, with their respective oc-
currences, are shown in Table 1 as the prevalence for the
other antigen combinations has not been established (Table
1).
39
The polymorphisms for the Ch and Rg epitopes as well
Table 1. Prevalence of Chido/Rodgers phenotypes in random samples
Chido
phenotype

Prevalence (%)
Rodgers
phenotype

Prevalence (%)
Ch:1,2,3 88.2 Rg:1,2 95.0
Ch:1,2,3 4.9 Rg:1,2 2.5
Ch:1,2,3 3.1 Rg:1,2 2.5
Ch:1,2,3 3.8
Ch:1,2,3 Rare
Ch:1,2,3 (proposed but Rg:1,2 (proposed but
Ch:1,2,3 may not exist) may not exist)
as the polymorphisms for the electrophoretic behavior of
the C4 protein confounded the early investigations on the
inheritance of the C4 genes. No attempt is going to be made
by this author to explain the reported phenotypic frequen-
cies of Ch and Rg antigens with the current understanding
of the multiple copy number variation of the C4 gene.
In early studies of the inheritance of Chido/Rodgers,
it appeared that the genes must be encoded by two closely
linked loci. The majority of persons were Chido-positive and
Rodgers-positive. Approximately 1.7 percent of those stud-
ied were Chido-negative, Rodgers-positive, and 3 percent
were Chido-positive, Rodgers-negative.
3,6
After the linkage
of Chido/Rodgers with HLA was established and the dis-
covery of C4 as the plasma protein carrier was made, family
studies detecting both serologic allelic forms and electro-
phoretic alleles seemed initially to be in agreement with
this model. It was proposed that null genes or duplicated
genes for C4A or C4B led to lack of Rodgers or Chido.
40
As methods and antisera improved, numerous C4 al-
leles were discovered and the results of a workshop to
defne and standardize nomenclature were published in
a joint paper by Mauff et al.
41
They described 14 determi-
nants for the C4A gene and 17 for the C4B gene. Genes that
express no C4A or C4B are designated with a *Q0 (mean-
ing quantity naught). In patients with complete defciency
of C4 (C4A*Q0/B*Q0), null alleles were thought to be at-
tributable to gene deletions or point mutations that cause
a frameshift in the DNA sequence, leading to defective or
no expression of the protein. One such example is a 2-bp
insertion into the sequence for codon 1213 at exon 29 of the
mutant C4A gene seen in both Caucasians and those of Af-
rican descent, but very rarely in those of Asian descent.
42
The origin of the complement system C3 and C4 pro-
teins predates the ability to make lymphocytes or antibod-
ies; the latter did not arise on the evolutionary scene until
cartilaginous fsh and sharks.
43
Homology between C3 and
C4 proteins indicates that the C4 gene may have evolved
from C3 as a duplicated gene.
44
The C4A gene is thought
to have arisen by gene duplication and mutation from C4B
because of its effciency in facilitating the clearance of IC.
28
The C4A or C4B gene consists of 41 exons that in ad-
dition to encoding for an acidic or basic isotype can also
exist as a long form of 20.6 kb or a short form of 14.2 kb.
45

The long form has an insertion of a human endogenous ret-
rovirus (HERV-K[C4]) into intron 9. The insertion of this
retroviral material is thought to have occurred after dupli-
cation of the ancestral mammalian C4A gene but before the
divergence of humans from the ancestors of other primates.
The reading frames of the retrovirus are all closed, but
there is still promoter activity and the HERV-K(C4) insert
is thought to be able to modulate the expression of the C4
gene.
46
The long form was originally thought to be a C4A
gene, but C4 genes can be C4A (long or short) or C4B (long
or short). However, most C4A genes have the HERV-K(C4)
insert and a greater proportion of C4B genes do not.
The genes for C4A and C4B were shown to be located
between HLA-DR and HLA-B loci on the short arm of chro-
mosome 6 along with C2 and factor B.
40,47
C4A and C4B
proteins are distinguished phenotypically by their migra-
tion patterns in electrophoretic gels and genetically by four
nucleotide substitutions that encoded changes at residues
1101 to 1106 where C4B encodes for LSPVIH and C4A en-
codes for PCPVLD at the protein level. C4A and C4B are
also distinguished phenotypically by their expression of ei-
ther the Chido or the Rodgers antigen, which are the result
of two amino acid substitutions at residues 1188 to 1191.
The Ch1 epitope has ADLR and the Rg1 has VDLL. As data
from family studies began to contradict a simple two-gene
model, it was realized that a haplotype could have one, two,
three, or four C4 genes and that the C4 genes were inher-
ited as part of a complex of genes.
48
The two-gene theory had
to give way to the current hypothesis of the inheritance of a
haplotype modular gene complex that can vary by C4 copy
number and is part of a larger gene complex in the MHC
locus. This model explains both the qualitative and quanti-
tative nature of the C4 gene expression as regards epitopes,
concentration of the C4 protein in the plasma, the frequency
of certain haplotypes, and the association of C4 copy number
variation with certain diseases. In fact, the C4 gene as part of
the MHC complex can serve as an example of how the genes
of an important biologic pathway evolve via the mechanics of
meiosis and through the pressure of natural selection to con-
serve important genes and to maximize the utility of poly-
morphism potential. Gene duplication, translocation, single-
nucleotide polymorphisms, insertions, and deletions all have
played a role in the plastic nature of the C4 gene.
Both forms of C4 are highly homologous but either
by SNP or by insertions or deletions display great qualita-
tive and quantitative diversity. The copy number variation
comes from whether an individual has inherited haplotypes
that have one, two, three, or four C4 genes.
49
This gene clus-
ter occurs as a complex in the MHC locus near the centromere
with fanking genes in the following order: frst an intact
RP gene (a serine/threonine kinase gene), an intact C4
gene (either C4A or C4B), then an intact CYP21B gene (cy-
tochrome P450 21-hydroxylase gene), and an intact TNXB
(tenascin-X protein). Approximately 17 percent of persons
of Northern European ancestry have this haplotype, termed
monomodular.
50
In 83 percent of those of Northern Euro-
pean ancestry the C4 gene complex can contain one, two,
or three duplicated gene clusters made of fragments or mu-
tations of the above fanking genes and an intact C4 gene.
The duplicated gene complex, called RCCX, occurs between
the C4 gene and the intact CYP21B gene and consists of a
CYP21 (usually a mutated form) fused to a 4.9-kb fragment
of TNXA fused to an RP2 0.91-kb fragment and an intact
C4 (A or B and long or short). Depending on the number of
RCCX modules duplicated, the complete C4 gene may be a
monomodular (no RCCX duplication), bimodular, trimod-
ular, or quadrimodular form.
R. Mougey
This modular gene duplication maximizes opportuni-
ties to conserve the C4 gene, whereas the opportunity for
recombination allows greater diversity of polymorphisms.
This great diversity should allow for a better response to
pathogens but can create opportunities for unequal cross-
ing over during meiosis. Thus low or high copy number
variations could be as important a risk factor for disease as
the actual C4 polymorphism.
C4 gene copy number studies done on healthy subjects
of different ages indicate that those individuals with a low
copy number for C4B may be less healthy or have a shorter
lifespan than those having more than two copies of the C4B
gene.
51
These studies found that in healthy individuals old-
er than 50 years, there was a lower incidence of low copy
number for the C4B gene when compared with the low copy
number gene frequency in younger healthy individuals.
C4 and Disease Associations
The association of C4 and a number of diseases has been
a fruitful area of investigation, and at the genetic level, the
frequency of C4 nulls or low copy variation shows a solid
association with autoimmune disease, particularly systemic
lupus erythematosus (SLE).
52
Although the varying effcien-
cies of complement function that result from partial or com-
plete defciency may have a profound effect on autoimmune
susceptibility, above-average C4 protein levels may also in-
crease risk. Once a loss of self-tolerance occurs, having high
levels of C4 may exacerbate the disease. Moreover, it cannot
be overstated that association is not the same as causation.
SLE can be thought of as a prototype for the intricacies of
the interaction of C4 proteins both in the development of
autoimmunity and in disease progression. Inheritance of a
C4A*Q0 null gene shows a greater risk in Caucasian and
African populations, but C4B*Q0 nulls are more frequent
in SLE patients of Asian descent.
53
SLE is a heterogeneous
disease that is characterized by the presence of multiple au-
toantibody specifcities that attack various tissues, result-
ing in tissue damage, which then presents more antigen to
the hosts immune system. It has become increasingly clear
that multiple genetic pathways are likely involved and that
C4 polymorphism, especially copy number variation, can
be a negative or a positive risk factor in the development
of autoimmune diseases.
54
This postulates that the function
of C4 in the initiation of autoimmunity may be related to
its role in presenting antigen to the host immune system,
which is separate from the role of complement in the in-
fammatory process once autoantibodies have been formed.
The current understanding of SLE genetics is that there are
distinctions in severity, risk, and clinical manifestation that
vary by ethnicity, geography, and sex with a greater preva-
lence in women and some non-Caucasian ethnic groups.
Although genetics undoubtedly plays a role, SLE may also
be associated with poverty and exposure to infectious dis-
eases.
55,56
Progress both in human genetics, particularly by
genome-wide association studies, and in the understand-
ing of biologic pathways has led to the discovery of more
than 20 different loci that show strong correlation to SLE.
54

Most of the genes identifed are involved in three kinds of
biologic processes. Minor variations in the genes encoding
the proteins of IC processing, production of interferon, or
signal transduction of immune cells can greatly increase the
susceptibility of an individual to SLE. It is to be expected
that more than one kind of trigger, such as pathogens, al-
lergens, or trauma, could then lead to disease.
Inasmuch as a major function of the C4 protein is par-
ti ci pati on i n i nnate i mmuni ty, there i s a strong
associ ation between defciencies of C4 and recurrent in-
fections; although this is less frequent than other comple-
ment defciencies.
57
Table 2 lists other reported diseases
and their possible association with C4 or complement.
In their paper, Atkinson and Frank
43
also warn that pa-
tients with complement C2- or C4-defcient states (either
acquired or inherited) have backup complement activation
mechanisms that must be considered in defning the role of
complement and a disease. For example, the immune re-
sponse of specifc antibody produced to a pathogen (antibody
Table 2. Disease associations with low levels of C4/complement
Date Disease Citation
2009 Schizophrenia Shi
58
2008 Role of complement in schizophrenia Mayilyan
59
2009 Chronic fatigue syndrome Sorenson
60
2008 Autism and missing or nonfunctional C4B
gene
Sweetehn
61
2008 Alzheimers and increased C4 deposition Zhou
62
2008 Increased morbidity/mortality associated
with low expression of C4B, smokers, and
stroke
Szilagyi
63
2007 Cardiovascular disease, smokers, and low
expression of C4B
Arason
64
2006 C4 nulls and adult rhinosinusitis Seppnen
65
2002 Systemic sclerosis Arason
66
1996 Insulin-dependent diabetes and C4 poly-
morphisms
Lhotta
67
1993 Graves disease Ratanachaiyavong
68
1991 C4A defciency and IgA nephropathy Wyatt
69
1990 Feltys syndrome Hillarby
70
1989 Psoriasis Wyatt
71
1989 Rheumatoid arthritis Fielder
72
1988 C4 allotypes in juvenile dermatomyositis Robb
73
1987 C4B2 and primary biliary cirrhosis Briggs
74
1984 Glomerulonephritis, Henoch-Schnlein
purpura
McLean
75
R. Mougey
excess) may be all that is needed to trigger more ancient
pathways to perform cell lysis in the absence of suffcient
complement proteins.
Summary
The C4 protein plays an important role in maintaining
health and, in some situations complicated by poor expres-
sion of the C4 protein, may lead to or exacerbate certain
diseases. The blood groups Chido and Rodgers are epitopes
on the C4 protein, and polymorphisms associated with
these epitopes may lead to the formation of antibodies to
the Chido or Rodgers antigens in transfused patients. Iden-
tifcation of anti-Ch or anti-Rg is still based on the antibody
neutralization with plasma from Ch-positive or Rg-positive
individuals and lack of reactivity with qualifed Ch-negative
or Rg-negative RBCs. These antibodies may be useful in
genetic studies of C4 polymorphisms or, in the case of C4-
defcient patients, a signal of the potential for serious ill-
nesses. The recognition of the extreme polymorphism of the
C4 gene and the gene complex RCCX should lead to more
insights in the understanding of disease risk and potential
treatment.
References
Harris JP, Tegoli J, Swanson J, Fisher N, Gavin J, No- 1.
ades J. A nebulous antibody responsible for crossmatch-
ing diffculties (Chido). Vox Sang 1967;12:1402.
Swanson J, Olson J, Azar M. Serological evidence that 2.
antibodies of Chido-York-Cs
a
specifcity are white cell
antibodies (abstract). Fed Proc 1971;30:248.
Middleton J, Crookston MC. Chido substance in plas- 3.
ma. Vox Sang 1972;23:25661.
Swanson JL. Laboratory problems associated with leu- 4.
kocyte antibodies. In: Seminar on recent advances in
immunohematology. Washington, DC: American Asso-
ciation of Blood Banks.1973 12155.
Middleton J, Crookston MC, Falk JA, et al. Linkage of 5.
Chido and HLA. Tissue Antigen 1974;4:36673.
Longster G, Giles CM. A new antibody specifcity, anti- 6.
Rg
a
, reacting with red cell and serum antigen. Vox Sang
1976;30:17580.
Nordhagen R, Olaisen B, Teisberg P, Gedde-Dahl T 7.
Jr. Association between the electrophoretically deter-
mined C4M haplotype product and partial inhibition of
anti-Ch
a
. J Immunogenet 1980;7:3016.
Giles CM. A new genetic variant for Chido. Vox Sang 8.
1984;46:14956.
Giles CM, Gedde-Dahl T Jr, Robson EB, et al. Rg 9.
a
(Rod-
gers) and the HLA region: linkage and association. Tis-
sue Antigens 1976;8:14349.
James J, Stiles P, Boyce F, Wright J. The HL-A type of 10.
Rg(a) individuals. Vox Sang 1976;30:21416.
ONeill GJ, Yang SY, Tegoli J, Berger R, Dupont B. 11.
Chido and Rodgers blood groups are distinct anti-
gen components of human complement C4. Nature
1978;273:66870.
Tilley CA, Romans DG, Crookston MC. Localization of 12.
Chido and Rodgers determinants to the C4d fragment
of human C4. Nature 1978;273:713 -5.
Judd WJ, Kraemer K, Moulds JJ. The rapid identifca- 13.
tion of Chido and Rodgers antibodies using C4d coated
red blood cells. Transfusion 1981;21:18992.
Giles CM, Hoffman M, Moulds M, Harris M, Delmasso 14.
A. Allo anti-Chido in a Ch-positive patient. Vox Sang
1987;52:12933.
Moulds JM, Nickells MW, Moulds JJ, et al. The C3b/ 15.
C4b receptor is recognized by the Knops, McCoy,
Swain-Langley, and York blood group sera. J Exp Med
1991;173:115963.
Rao N, Ferguson DJ, Lee SF, Telen MJ. Identifcation 16.
of human erythrocyte blood group antigens of C3b/C4b
receptor. J Immunol 1991;146:35017.
Lambin P, LePennec PY, Hauptmann G, et al. Adverse 17.
transfusion reactions associated with a precipitating
anti-C4 antibody of anti-Rodgers specifcity. Vox Sang
1984;47:2429.
Westhoff CM, Sipherd BD, Wylie DE, Toalson LD. 18.
Severe anaphylactic reactions following transfusions
of platelets to a patient with anti-Ch. Transfusion
1992;32:5769.
Strupp A, Cash K, Uehlinger J. Diffculties in identi- 19.
fying antibodies in the Dombrock blood group sys-
tem in multiply alloimmunized patients, Transfusion
1998;38:10225.
Awdeh ZL, Alper CA. Inherited structural polymor- 20.
phism of the fourth component of human complement.
Proc Natl Acad Sci U S A 1980;77:357680.
Rittner C, Giles CM, Roos MH, Demant P, Mollenhauer 21.
E. Genetics of human C4 polymorphism: detection and
segregation of rare and duplicated haplotypes. Immu-
nogenetics 1984;19:32133.
Roos MH, Giles CM, Demant P, Mollenhuuer E, Ritt- 22.
ner C. Rodgers (Rg) and Chido (Ch) determinants on
human C4: characterization of two C4B5 subtypes, one
of which contains Rg and Ch determinants. J Immunol
1984;133:263440.
Yu CY. The complete exon-intron structure of a human 23.
complement component C4A gene. DNA sequences,
polymorphisms and linkage to 21-hydroxylase gene. J
Immunol 1991;146:105766.
Yu CY, Campbell RD, Porter RR. A structural model for 24.
the location of the Rodgers and Chido antigenic deter-
minants and their correlation with the human comple-
ment components C4A/C4B isotypes. J Immunogenet
1988;27:399405.
Belt KT, Carroll MC, Porter RR. The structural basis of 25.
the multiple forms of human complement component
C4. Cell 1984;36:90714.
Blanchong CA, Chung EK, Rupert KL, et al. Genetic, 26.
structural and functional diversities of human comple-
ment components C4A and C4B and their mouse homo-
logues Slp and C4. Int Immunopharmacol 2001;1:365
92.
Law SK, Lichtenberg NA, Levine RP. Covalent binding 27.
and hemolytic activity of complement proteins. Proc
Natl Acad Sci U S A 1980;77:71948.
Dodds AW, Ren X-D, Willis AC, Law SKA. The reac- 28.
tion mechanism of the internal thioester in the human
complement component C4. Nature 1996;379:1779.
Carroll MC, Fathallah DM, Bergamaschini L, Alicot EM, 29.
Isenman DE. Substitution of a single amino acid (as-
partic acid for histidine) converts the functional activity
of human complement C4B to C4A. Proc Natl Acad Sci
U S A 1990;87:686872.
Colton HR. Genetics and synthesis of components of 30.
the complement system. In: Ross GD, ed. Immunobi-
ology of the complement system. An introduction for
research and clinical medicine. New York: Academic
Press, 1986:1746.
Uko G, Christiansen FT Dawkins RL, McCann VJ. 31.
Reference ranges for serum C4 concentrations in sub-
jects with and without C4 null alleles. J Clin Pathol
1986;39:5736.
Moulds JM, Arnett FC, Giles CM, Hamilton RG. A novel 32.
immunoassay for the quantifcation of human C4 gene
products. Complement Infamm 1990;7:93101.
Chung EK, Yang Y, Rupert KL, et al. Determining the 33.
one, two, three, or four long and short loci human com-
plement C4 in a major histocompatibility complex hap-
lotype encoding C4A or C4B proteins. Am J Hum Genet
2002;71:81022.
Fukuoka Y, Zia HZ, Sanchez-Munoz LB, Dellinger AL, 34.
Escribano L, Schwartz B. Generation of anaphylatox-
ins by human -tryptase form C3, C4, C5. J Immunol
2008;180:630716.
Richani K, Soto E, Romero R, et al. Normal pregnancy is 35.
characterized by systemic activation of the complement
system. J Matern Fetal Neonate Med 2005;17:23945.
Gigli I, Fugita T, Nussenzweig V. Modulation of the 36.
classical pathway C3 convertase by plasma proteins C4
binding protein and C3b inactivator. Proc Natl Acad Sci
U S A 1979;76:6596600.
Gatenby PA, Barbosa JE, Lachmann PJ. Differences 37.
between C4A and C4B in the handling of immune com-
plexes: the enhancement of CR1 binding is more impor-
tant than the inhibition of immunoprecipitation. Clin
Exp Immunol 1990;79:15863.
Manderson AP, Pickering MC, Botto M, Walport MJ, 38.
Parrish CR. Continual low level activation of the classi-
cal complement pathway. J Exp Med 2001;194:74756.
Giles CM. An update on Rodgers and Chido, the anti- 39.
genic determinants of human C4. Immunohematology
1989;5:16.
Carroll MC, Palsdottir A, Belt KT, Porter RR. Deletion 40.
of complement C4 and 21-hydroxylase genes in the
HLA class III region. EMBO J 1985;4:254752.
Mauff G, Alper CA, Dawkins RL, et al. C4 nomenclature 41.
statement. Complement Infamm 1991;7:2618.
Barba G, Rittner C, Schneider PM. Genetic basis of 42.
human complement C4A defciency, detection of a
point mutation leading to nonexpression. J Clin Invest
1993;91:16816.
Atkinson JP, Frank MM. Bypassing complement: evo- 43.
lutionary lessons and future implications. J Clin Invest
2006;116:12158.
Martinez OP, Longman-Jacobsen N, Davies R, et al. Ge- 44.
netics of human complement component C4 and evolu-
tion of the central MHC. Front Biosci 2001;6:D904
13.
Dangel AW, Mendoza AR, Baker BJ, et al. The dichoto- 45.
mous size variation of human complement C4 gene is
mediated by a novel family of endogenous retroviruses
which also establishes species-specifc genomic pat-
terns among Old World primates. Immunogenetics
1994;40:42536.
Mack M, Bender K, Schneider PM. Detection of retro- 46.
viral antisense transcripts and promoter activity of the
HERV-K(C4) insertion in the MHC class III region. Im-
munogenetics 2004;56:32132.
Yu CY, Campbell RD. Defnitive RFLPS to distinguish 47.
between the human complement C4/C4B isotypes
and the major Rodgers/Chido determinants: applica-
tion to the study of C4null alleles. Immunogenetics
1987;25:38390.
Yu CY, Yang Z, Blanchong CA, Miller W. The human 48.
and mouse MHC class III region: a parade of the cen-
tromeric segment with 21 genes. Immunol Today
2000;21:3208.
Chung EK, Yang Y, Rupert KL, et al. Determining the 49.
one, two, three or four long and short loci of human
complement C4 in a major histocompatibility complex
haplotype encoding C4A or C4B proteins. Am J Hum
Genet 2002;71:81022.
Wu YL, Yang Y, Chung EK, et al. Phenotypes, geno- 50.
types and disease susceptibility associated with gene
copy number variations: complement C4 CNVs in Eu-
ropean American healthy subjects and those with sys-
temic lupus erythematosus. Cytogenet Genome Res
2008;123:13141.
Szilagyi A, Fust G. Diseases associated with low copy 51.
number of the C4B gene encoding C4, the fourth
component of complement. Cytogenet Genome Res
2008;123:11830.
R. Mougey
Blanchong CA, Zhou B, Rupert KL, et al. Defciencies 52.
of human complement component C4A and C4B and
heterozygosity in length variants of RP-C4-CYP21-TNX
(RCCX) modules in Caucasians: the load of RCCX ge-
netic diversity on major histocompatibility complex-
associated disease. J Exp Med 2000;191:218396.
Yang Y, Chung EK, Zhou B, et al. The intricate role of 53.
complement component C4 in human systemic lupus
erythematosus. Curr Dir Autoimmun. 2004;7:98132.
Harley IT, Kaufman KM, Langfeld CD, Harley JB, Kelly 54.
JA. Genetic susceptibility to lupus: new insights from
fne mapping and genome-wide association studies.
Nat Rev Genet 2009;10:28590.
Crosslin KL, Wiginton KL. The impact of race and eth- 55.
nicity on disease severity in systemic lupus erythemato-
sus. Ethn Dis 2009;19:3017.
Hiraki LT, Benseler SM, Tyrrell PN, Harvey E, Herbert D, 56.
Silverman ED. Ethnic differences in pediatric systemic
lupus erythematosus. J Rheumatol 2009;36:253946.
Figueroa JE, Densen P. Infectious diseases associ- 57.
ated with complement defciencies. Clin Microbiol Rev
1991;4:35995.
Shi J, Levinson DF, Duan J, et al. Common variants on 58.
chromosome 6p22.1 are associated with schizophrenia.
Nature 2009;460:7537.
Mayilyan LR, Weinberger DR, Sim RB. The comple- 59.
ment system in schizophrenia. Drug News Perspect
2008;21:20010.
Sorensen B, Jones JF, Vernon SD, Rajeevan MS. Tran- 60.
scription control of complement activation in an ex-
ercise model of chronic fatigue syndrome. Mol Med
2009;15:3442.
Sweeten TL, Odell DW, Odell JD, Torres AR. C4B null 61.
alleles are not associated with genetic polymorphisms
in the adjacent gene CYP21A2 in autism. BMC Med
Genet 2008;9:17.
Zhou J, Fonseca MI, Pisalyaput K, Tenner AJ. Comple- 62.
ment C3 and C4 expression in C1q suffcient and de-
fcient mouse models of Alzheimers disease. J Neuro-
chem 2008;106:208092.
Szilagyi A, Fust G. Diseases associated with the low 63.
copy number of the C4B gene encoding C4, the fourth
component of complement. Cytogenet Genome Res
2008;123:11830.
Arason GJ, Kramer J, Blasko B, et al. Smoking and a 64.
complement gene polymorphism interact in promoting
cardiovascular disease morbidity and mortality. Clin
Exp Immunol 2007;149:1328.
Seppnen M, Suvilehto J, Lokki M-L, et al. Immuno- 65.
globulins and complement factor C4 in adult rhinosi-
nusitis. Clin Exp Immunol 2006;145:21927.
Arason GJ, Geirsson AJ, Kolka R, Vikingsdottir TH, 66.
Valdimarrson H. Defciency of complement dependent
prevention of immune precipitation in systemic sclero-
sis. Ann Rheum Dis 2002;61:25760.
Lhotta K, Auinger M, Kronenberg F, Irsigler K, Konig P. 67.
Polymorphism of complement C4 and susceptibility to
IDDM and microvascular complications. Diabetes Care
1996;19:535.
Ratanachaiyavong S, Demaine AG, Campbell RD, 68.
McGregore AM. Heat shock protein 70 (HS70) and
complement C4 genotypes in patients with hyperthy-
roid Graves disease. Clin Exp Immunol 1991;84:48
52.
Wyatt RJ, Julian BA, Woodford SY. C4A defciency and 69.
poor prognosis in patients with IgA nephropathy. Clin
Nephrol 1991;36:15.
Hillarby MC, Strachan T, Grennan DM. Molecular 70.
characterisation of C4 null alleles found in Feltys syn-
drome. Ann Rheum Dis 1990;49:7637.
Wyatt RJ, Wang C, Hudson EC, et al. Complement 71.
phenotypes in patients with psoriasis. Hum Hered
1989;39:32732.
Fielder AH, Ollier W Lord DK, et al. HLA class III hap- 72.
lotypes in multicase rheumatoid arthritis families. Hum
Immunol 1989;25:7585.
Robb SA, Fielder AHL, Saunders CEL. C4 complement 73.
allotypes in juvenile dermatomyositis. Hum Immunol
1988;22:318.
Briggs DC, Donaldson PT, Hayes P, et al. A major histo- 74.
compatibility complex class III allotype (C4B2) associ-
ated with primary biliary cirrhosis (PBC). Tissue Anti-
gens 1987;29:1415.
McLean RH, Wyatt RJ, Julian BA. Complement pheno- 75.
types in glomerulonephritis. Increased frequency of ho-
mozygous null C4 phenotypes in IgA nephropathy and
Henoch-Schnlein purpura. Kidney Int 1984;26:855
60.

Ruth Mougey MT(ASCP)SBB, President, Mougey, Incor-
porated, 1705 Highland Avenue, Carrollton KY 41008.
Immunohematology is on the Web!
www.redcross.org/immunohematology
For more information, send an e-mail message
to immuno@usa.redcross.org
Important Notice About Manuscripts
for
Immunohematology
Please e-mail all manuscripts for consideration to
immuno@usa.redcross.org
In Memoriam: Marie Cutbush Crookston
19202009
D. Mallory
Marie Cutbush Crookston, a dynamic person and ex-
ceptional scientist, was born in 1920 in Victoria, Australia.
She received a Bachelor of Science degree from the Univer-
sity of Melbourne, and, in 1947, she set sail for England.
There, she worked for 10 years with Dr. P. L. Mollison at
the Medical Research Council Blood Transfusion Research
Unit and contributed to the writing of the frst edition of Dr.
Mollisons Blood Transfusion in Clinical Medicine.
In 1957, she married Dr. John Crookston, a hematolo-
gist from Toronto, who was in London working for Dr. J. V.
Dacie. Giving up her studies for a PhD, she moved to To-
ronto to raise two sons and to continue to work in the feld
of immunohematology. In 1964 she joined the University
of Toronto, Department of Pathology, where she directed
research, and in 1978 she became an associate professor.
She also served as immunohematologist in the Department
of Hematology at the Toronto General Hospital and acted
as a consultant for the Blood Transfusion Laboratory until
she retired in 1986.
Maries contributions to immunohematology are out-
standing and have made a permanent impact on the feld.
She and her late husband John helped transform the edu-
cational level and direction of immunohematology in the
1960s when they returned to Toronto from London. There,
they helped to organize the Ontario Antibody Club, which
was active for 25 years.
Maries studies on conversion of incomplete to complete
antibodies by chemical modifcation were used by commer-
cial companies to make chemically modifed Rh antiserum.
In addition, she was known for her discoveries of anti-Fy
a

and anti-Lu
b
and her early work in exchange transfusion
of the newborn, long-term preservation of RBCs by freez-
ing, and linkage of Chido and HLA. She and John Crook-
ston also coined the term HEMPAS to describe an inherited
chronic hemolytic anemia with multinucleated cells in the
bone marrow associated with a positive acidifed serum test
(modifed Hams solution).
Everyone who knew Marie and read any of her 65 pub-
lished papers remembers her wonderful ability to write in
plain intelligible English, or as she would say, deathless
prose, and how she kept strictly to proper blood group ter-
minology. Her papers were always accepted as submitted.
She was on the editorial boards of Immunohematology,
Journal of Blood Group Serology and Education, Transfu-
sion, and Vox Sanguinis.
These are some of her awards:
The Karl Landsteiner Memorial Award in 1976 with Dr.
Eloise Giblett from the American Association of Blood
Banks, awarded for her discoveries of anti-Fy
a
, anti-
Lu
b
, and anti-HEMPAS: for her studies of the relation-
ship between blood-group antibody activity in vitro and
in vivo: for her description of red cell uptake of Le
a
and
Le
b
antigens from plasma; for her investigation of the
Chido antigen in plasma; and for the detection of the
linkage between Ch and HLA. (quotation from citation
on award)
The Nuffeld Foundation Award in 1952 with Dr. Her-
mann Lehmann for studying the blood groups of tribes
in the Nilgiri Hills of South India.
The Buchanan Award in 1980 from the Canadian Soci-
ety of Transfusion Medicine.
The Canadian Blood Services Award in 2002 for Life-
time Achievement.
Marie always found the time to help students and scien-
tists who needed directions with their studies or help with
their manuscripts. She was a wonderful, caring person who
was an enthusiastic, intelligent, and brilliant scientist and
one who will be missed by her many friends and colleagues
around the world.
In Memoriam: Eloise Giblett 19212009
D. Mallory
Eloise Elo Giblett was born in Tacoma, Washington,
in 1921 and, in 1931, the family moved to Spokane. Elo de-
veloped a keen interest in music at an early age, taking pi-
ano, dance, and violin, the instrument on which she would
focus, becoming concertmaster of her high school orches-
tra. She put aside music, turned to science, and received a
BS in bacteriology from the University of Washington (U
of W) in 1942. World War II had started, so Elo joined the
WAVES in 1944, became a medical laboratory technician,
and served until 1946. She returned to the U of W and re-
ceived her MS degree in microbiology in 1947 and then, in
1951, graduated frst in her class in the second class to grad-
uate from the new medical school at the U of W.
During the medical internship and hematology fellow-
ship at U of W, her interest in human genetics was encour-
aged. In 1955, she became the frst full-time physician at
King County Blood Bank (now the Puget Sound Blood Cen-
ter) as the Associate Director of the Typing and Crossmatch
Laboratory and was given a 6-month sabbatical with Dr.
Patrick Mollison at the Blood Transfusion Research Unit in
London, England. There, she studied the Lewis blood group
system, measured the rate of tagged RBC destruction, be-
came acquainted with the clinical potential of the Coombs
antiglobulin test, and mastered the tests used at that time
in Dr. Mollisons laboratory. When she returned to Seattle,
she began to use the antiglobulin test to fnd RBC antibod-
ies and discovered anti-V and anti-Js
a
.
Her numerous studies of the polymorphisms of hap-
toglobin and transferrins and her interest in RBC, plasma,
and protein polymorphisms resulted in the renowned 1969
publication of her book, Genetic Markers in Human Blood.
In addition, she discovered that two forms of inherited im-
munodefciency disease were caused by defciencies of the
enzymes adenosine deaminase and purine nucleoside phos-
phorylase.
In 1979, Dr. Giblett became the Director of the Puget
Sound Blood Center and remained so until she retired in
1987. Upon her retirement, she became Director Emeritus
of the Blood Center and Professor Emeritus of the U of W
Medical School. After 40 years, she returned to the violin
and even helped found a new music school.
Dr. Giblett was president of the American Society of
Human Genetics (1973); a member of the Advisory Board of
the American Society of Hematology (19801986); a mem-
ber of the Editorial Board of Blood, Transfusion, and The
American Journal of Human Genetics (among a few); and
chairperson of the Enzyme Nomenclature Committee of the
International Workshop on Human Gene Mapping.
She received many honors during her lengthy career, includ-
ing two from the American Association of Blood Banks:
1975Emily Cooley Lecture Award (American Associa-
tion of Blood Banks)
1976Karl Landsteiner Memorial Award (American
Association of Blood Banks)
1978The Philip S. Levine Award (American Society of
Clinical Pathology)
1980Election, National Academy of Science
1987Distinguished Alumna Award (University of
Washington)
Dr. Giblett has left a legacy of original scientifc discov-
eries, a book that was groundbreaking for the feld of im-
munohematology, and colleagues who were scientifcally
mentored and generously helped by her excellent editorial
skills. Those who knew her will remember a very interesting
and giving human being.
Masters (MSc) in Transfusion and Transplantation Sciences
at
The University of Bristol, England
Applications are invited from medical or science graduates for the Master of Science (MSc)
degree in Transfusion and Transplantation Sciences at the University of Bristol. The course
starts in October 2010 and will last for 1 year. A part-time option lasting 2 or 3 years is also
available. There may also be opportunities to continue studies for PhD or MD following the
MSc. The syllabus is organized jointly by The Bristol Institute for Transfusion Sciences and the
University of Bristol, Department of Pathology and Microbiology. It includes:
Scientifc principles of transfusion and transplantation
Clinical applications of these principles
Practical techniques in transfusion and transplantation
Principles of study design and biostatistics
An original research project
Application can also be made for Diploma in Transfusion and Transplantation Science or a
Certifcate in Transfusion and Transplantation Science.
The course is accredited by the Institute of Biomedical Sciences.
Further information can be obtained from the Web site:
http://www.blood.co.uk/ibgrl/MscHome.htm
For further details and application forms please contact:
Dr Patricia Denning-Kendall
University of Bristol, Paul OGorman Lifeline Centre, Department of Pathology and
Microbiology, Southmead Hospital, Westbury-on-Trym, Bristol BS10 5NB, England
Fax +44 1179 595 342, Telephone +44 1779 595 455, e-mail: p.a.denning-kendall@bristol.
ac.uk.
Announcements
Diagnostic testing for:
Neonatal alloimmune thrombocytopenia (NAIT)
Posttransfusion purpura (PTP)
Refractoriness to platelet transfusion
Heparin-induced thrombocytopenia (HIT)
Alloimmune idiopathic thrombocytopenia purpura (AITP)
Medical consultation available
Test methods:
GTI systems tests
detection of glycoprotein-specifc platelet antibodies
detection of heparin-induced antibodies (PF4 ELISA)
Platelet suspension immunofuorescence test (PSIFT)
Solid phase red cell adherence (SPRCA) assay
Monoclonal immobilization of platelet antigens (MAIPA)
Molecular analysis for HPA-1a/1b
For further information, contact
Platelet Serology Laboratory (215) 451-4205
Maryann Keashen-Schnell (215) 451-4041 offce
mschnell@usa.redcross.org
Sandra Nance (215) 451-4362
snance@usa.redcross.org
American Red Cross Blood Services
Musser Blood Center
700 Spring Garden Street
Philadelphia, PA 19123-3594
CLIA licensed
National Neutrophil Serology Reference
Laboratory
Our laboratory specializes in granulocyte antibody detection
and granulocyte antigen typing.
Indications for granulocyte serology testing include:
Alloimmune neonatal neutropenia (ANN)
Autoimmune neutropenia (AIN)
Transfusion related acute lung injury (TRALI)
Methodologies employed:
Granulocyte agglutination (GA)
Granulocyte immunofuorescence by fow cytometry (GIF)
Monoclonal antibody immobilization of neutrophil antigens
(MAINA)
TRALI investigations also include:
HLA (PRA) Class I and Class II antibody detection
For further information contact:
Neutrophil Serology Laboratory (651) 291-6797
Randy Schuller(651) 291-6758
schullerr@usa.redcross.org
Neutrophil Serology Laboratory
100 South Robert Street
St. Paul, MN 55107
CLIA licensed
National Platelet Serology Reference Laboratory
Monoclonal antibodies available at no charge
The New York Blood Center has developed a wide range of monoclonal antibodies (both murine and humanized) that are
useful for donor screening and for typing RBCs with a positive DAT. These include anti-A1, -M, -s, -U, -D, -Rh17, -K, -k,
-Kp
a
, -Js
b
, -Fy
a
, -Fy3, -Fy6, Wr
b
, -Xg
a
, -CD99, -Do
b
, -H, -Ge2, -Ge3, -CD55 (both SCR2/3 and SCR4), -Ok
a
, -I, and anti-
CD59. Most of the antibodies are murine IgG and require the use of anti-mouse IgG for detection (anti-K, k, and -Kp
a
).
Some are directly agglutinating (anti-A1, -M, -Wr
b
and -Rh17) and a few have been humanized into the IgM isoform (anti-
Js
b
). The antibodies are available at no charge to anyone who requests them. Please visit our Web site for a complete list of
available monoclonal antibodies and the procedure for obtaining them.
For additional information, contact: Gregory Halverson, New York Blood Center, 310 East 67th Street, New York, NY
10021; e-mail: ghalverson@nybloodcenter.org; phone: (212) 570-3026; fax: (212) 737-4935; or visit the web site at http://
www.nybloodcenter.org >research >immunochemistry >current list of monoclonal antibodies available.
Specialist in Blood Bank (SBB) Program
The Department of Transfusion Medicine, National Institutes of Health, is accepting applications for its 1-year Specialist
in Blood Bank Technology Program. Students are federal employees who work 32 hours/week. This program introduces
students to all areas of transfusion medicine, including reference serology, cell processing, HLA, and infectious disease
testing. Students also design and conduct a research project. NIH is an Equal Opportunity Organization. Application
deadline is December 31, 2010, for the July 2011 class. See www.cc.nih.gov/dtm > education for brochure and application.
For further information contact Karen M. Byrne at (301) 451-8645 or KByrne@mail.cc.nih.gov.
Announcements, cont., and advertisements
Advertisements, cont.
Reference and Consultation Services
Antibody identifcation and problem resolution
HLA-A, B, C, and DR typing
HLA-disease association typing
Paternity testing/DNA
For information, contact
Mehdizadeh Kashi at (503) 280-0210, or write to:
Pacifc Northwest Regional Blood Services
ATTENTION: Tissue Typing Laboratory
American Red Cross
3131 North Vancouver
Portland, OR 97227
CLIA licensed, ASHI accredited
IgA/Anti-IgA Testing
IgA and anti-IgA testing is available to do the
following:
Identify IgA-defcient patients
Investigate anaphylactic reactions
Confrm IgA-defcient donors
Our ELISA for IgA detects protein to 0.05 mg/dL.
For additional information contact Cindy Flickinger at
(215) 451-4909, or e-mail: fickingerc@usa.redcross.org,
or write to:
Musser Blood Center
ATTN: Cindy Flickinger
CLIA licensed
Donor IgA Screening
Effective tool for screening large volumes of
donors
Gel diffusion test that has a 15-year proven track
record:
Approximately 90 percent of all donors
identifed as IgA defcient by are confrmed by
the more sensitive testing methods
For additional information, call Kathy Kaherl at:
(860)678-2764, e-mail: kaherlk@usa.redcross.org
or write to:
Reference Laboratory
American Red CrossBiomedical Services
Connecticut Region
209 Farmington Ave.
Farmington, CT 06032
CLIA licensed
National Reference Laboratory
for Blood Group Serology
Immunohematology Reference Laboratory
AABB, ARC, New York State, and CLIA licensed
(215) 451-4901 24-hr. phone number
(215) 451-2538 Fax
American Rare Donor Program
(215) 451-4900 24-hr. phone number
(215) 451-2538 Fax
ardp@usa.redcross.org
Immunohematology
(215) 451-4902 Phone, business hours
(215) 451-2538 Fax
immuno@usa.redcross.org
Quality Control of Cryoprecipitated-AHF
(215) 451-4903 Phone, business hours
(215) 451-2538 Fax
Ordering Information
The book, which costs $25, can be
ordered in two ways:
Order online from the publisher
at: www.sbbpocketbook.com
Order from the authors, who
will sign the book. Send a check,
made payable to New York Blood
Center and indicate Pocket-
book on the memo line, to:
Marion Reid
Laboratory of Immunochemistry
New York Blood Center
310 East 67th Street
New York, NY 10065
Please include the recipients
complete mailing address.
Blood Group
Antigens &
Antibodies
A guide to clinical relevance
& technical tips
by Marion E. Reid & Christine Lomas-Francis
Pocketbook Education Fund
The authors are using royalties generated from the sale of
this pocketbook for educational purposes to mentor people
in the joys of immunohematology as a career. They will
accomplish this in the following ways:
Sponsor workshops, seminars, and lectures
Sponsor students to attend a meeting
Provide copies of the pocketbook
(See www.sbbpocketbook.com for details to apply for funds)
This compact pocketbook from the authors of the Blood
Group Antigen FactsBook is a must for anyone who is involved
in the laboratory or bedside care of patients with blood
group alloantibodies.
The book contains clinical and technical information about
the nearly 300 ISBT recognized blood group antigens and
their corresponding antibodies. The information is listed in
alphabetical order for ease of findingeven in the middle of
the night. Included in the book is information relating to:
Clinical significance of antibodies in transfusion and HDN.
Number of compatible donors that would be expected
to be found in testing 100 donors. Variations in different
ethnic groups are given.
Characteristics of the antibodies and optimal technique(s)
for their detection.
Technical tips to aid their identification.
Whether the antibody has been found as an autoantibody.
Additional Information can be found by visiting the following Web sites: www.ascp.org, www.caahep.org and www.aabb.org Revised August 2007
Becoming a Specialist in Blood Banking (SBB)
What is a certifed Specialist in Blood Banking (SBB)?
Someone with educational and work experience qualifcations who successfully passes the American Society for Clinical Pathology (ASCP)
board of registry (BOR) examination for the Specialist in Blood Banking.
This person will have advanced knowledge, skills, and abilities in the feld of transfusion medicine and blood banking.
Individuals who have an SBB certifcation serve in many areas of transfusion medicine:
Serve as regulatory, technical, procedural and research advisors
Perform and direct administrative functions
Develop, validate, implement, and perform laboratory procedures
Analyze quality issues preparing and implementing corrective actions to prevent and document issues
Design and present educational programs
Provide technical and scientifc training in blood transfusion medicine
Conduct research in transfusion medicine
Who are SBBs?
Supervisors of Transfusion Services Managers of Blood Centers LIS Coordinators Educators
Supervisors of Reference Laboratories Research Scientists Consumer Safety Offcers
Quality Assurance Offcers Technical Representatives Reference Lab Specialist
Why be an SBB?
Professional growth Job placement Job satisfaction Career advancement
How does one become an SBB?
Attend a CAAHEP-accredited Specialist in Blood Bank Technology Program OR
Sit for the examination based on criteria established by ASCP for education and experience
However: In recent years, a greater percentage of individuals who graduate from CAAHEP-accredited programs pass the SBB exam.

Conclusion:
The BEST route for obtaining an SBB certifcation is
to attend a CAAHEP-accredited Specialist in Blood Bank Technology Program
Contact the following programs for more information:

Program Contact Name Phone Contact Email Contact Website On site
or On line
Program
Walter Reed Army Medical Center William Turcan 202-782-6210 William.Turcan@NA.AMEDD.ARMY.
MIL
www.militaryblood.dod.mil On site
American Red Cross, Southern California Region Michael Coover 909-859-7496 CooverM@usa.redcross.org none On site
ARC-Central OH Region Joanne Kosanke 614-253-2740 x 2270 kosankej@usa.redcross.org none On site
Blood Center of Southeastern Wisconsin Lynne LeMense 414-937-6403 Lynne.Lemense@bcw.edu www.bcw.edu On site
Community Blood Center/CTS Dayton, Ohio Nancy Lang 937-461-3293 nlang@cbccts.org http://www.cbccts.org/education/
sbb.htm
On line
Gulf Coast School of Blood Bank Technology Clare Wong 713-791-6201 cwong@giveblood.org www.giveblood.org/education/
distance/htm
On line
Hoxworth Blood Center, Univ. of Cincinnati Susan Wilkinson 513-558-1275 susan.wilkinson@uc.edu www.hoxworth.org On site
Indiana Blood Center Jayanna Slayten 317-916-5186 jslayten@indianablood.org www.indianablood.org On line
Johns Hopkins Hospital Jan Light 410-955-6580 jlight5@jhmi.edu http://pathology2.jhu/
department/divisions/trnafusion/
sbb.cfm
On site
Medical Center of Louisiana Karen Kirkley 504-903-3954 kkirkl@lsuhsc.edu none On site
NIH Clinical Center Dept.. of Transfusion Medicine Karen Byrne 301-496-8335 Kbyrne@mail.cc.nih.gov www.cc.nih.gov/dtm On site
Rush University Veronica Lewis 312-942-2402 Veronica_Lewis@rush.edu www.rushu.rush.edu/health/dept.
html
On line
Transfusion Medicine Center at Florida Blood
Services
Marjorie Doty 727-568-5433 x 1514 mdoty@fbsblood.org www.fbsblood.org On line
Univ. of Texas Health Science Center at San Antonio Linda Myers 210-731-5526 lmyers@bloodntissue.org www.uthscsa.edu On site
Univ. of Texas Medical Branch at Galveston Janet Vincent 409-772-3055 jvincent@utmb.edu www.utmb.edu/sbb On line
Univ. of Texas SW Medical Center Barbara Laird-
Fryer
214-648-1785 barbara.fryer@UTSouthwestern.edu http://telecampus.utsystem.edu On line
I M M U N O H E M A T O L O G Y, V O L U M E 2 2 , N U M B E R 4 , 2 0 0 6 215
I. GENERAL INSTRUCTIONS
Before submitting a manuscript, consult current issues of
Immunohematology for style. Double-space throughout the manuscript.
Number the pages consecutively in the upper right-hand corner, beginning
with the title page.
II. SCIENTIFIC ARTICLE, REVIEW, OR CASE REPORT WITH
LITERATURE REVIEW
A. Each component of the manuscript must start on a new page in the
following order:
1. Title page
2. Abstract
3. Text
4. Acknowledgments
5. References
6. Author information
7. Tables
8. Figures
B. Preparation of manuscript
1. Title page
a. Full title of manuscript with only first letter of first word capitalized
(bold title)
b. Initials and last name of each author (no degrees; all CAPS), e.g., M.T.
JONES, J.H. BROWN, AND S.R. SMITH
c. Running title of 40 characters, including spaces
d. Three to ten key words
2. Abstract
a. One paragraph, no longer than 300 words
b. Purpose, methods, findings, and conclusion of study
3. Key words
a. List under abstract
4. Text (serial pages): Most manuscripts can usually, but not necessarily,
be divided into sections (as described below). Survey results and
review papers may need individualized sections
a. Introduction
Purpose and rationale for study, including pertinent background
references
b. Case Report (if indicated by study)
Clinical and/or hematologic data and background serology/molecular
c. Materials and Methods
Selection and number of subjects, samples, items, etc. studied and
description of appropriate controls, procedures, methods, equipment,
reagents, etc. Equipment and reagents should be identified in
parentheses by model or lot and manufacturers name, city, and state.
Do not use patients names or hospital numbers.
d. Results
Presentation of concise and sequential results, referring to pertinent
tables and/or figures, if applicable
e. Discussion
Implication and limitations of the study, links to other studies; if
appropriate, link conclusions to purpose of study as stated in
introduction
5. Acknowledgments: Acknowledge those who have made substantial
contributions to the study, including secretarial assistance; list any grants.
6. References
a. In text, use superscript, Arabic numbers.
b. Number references consecutively in the order they occur in the text.
7. Tables
a. Head each with a brief title; capitalize the first letter of first word (e.g.,
Table 1. Results of . . .) use no punctuation at the end of the title.
b. Use short headings for each column needed and capitalize first letter
of first word. Omit vertical lines.
c. Place explanation in footnotes (sequence: *, , , , , **, ).
8. Figures
a. Figures can be submitted either by e-mail or as photographs (5 7
glossy).
b. Place caption for a figure on a separate page (e.g. Fig. 1 Results of. . . ),
ending with a period. If figure is submitted as a glossy, place first
authors name and figure number on back of each glossy submitted.
c. When plotting points on a figure, use the following symbols if
possible: .
9. Author information
a. List first name, middle initial, last name, highest degree, position held,
institution and department, and complete address (including ZIP
code) for all authors. List country when applicable.
III. EDUCATIONAL FORUM
A. All submitted manuscripts should be approximately 2000 to 2500
words with pertinent references. Submissions may include:
1. An immunohematologic case that illustrates a sound investigative
approach with clinical correlation, reflecting appropriate collaboration
to sharpen problem solving skills
2. Annotated conference proceedings
B. Preparation of manuscript
1. Title page
a. Capitalize first word of title.
b. Initials and last name of each author (no degrees; all CAPs)
2. Text
a. Case should be written as progressive disclosure and may include the
following headings, as appropriate
i. Clinical Case Presentation: Clinical information and differential
diagnosis
ii. Immunohematologic Evaluation and Results: Serology and
molecular testing
iii. Interpretation: Include interpretation of laboratory results,
correlating with clinical findings
iv. Recommended Therapy: Include both transfusion and
nontransfusion-based therapies
v. Discussion: Brief review of literature with unique features of this
case
vi. Reference: Limited to those directly pertinent
vii. Author information (see II.B.9.)
viii. Tables (see II.B.7.)
IV. LETTER TO THE EDITOR
A. Preparation
1. Heading (To the Editor)
2. Title (first word capitalized)
3. Text (written in letter [paragraph] format)
4. Author(s) (type flush right; for first author: name, degree, institution,
address [including city, state, Zip code and country]; for other authors:
name, degree, institution, city and state)
5. References (limited to ten)
6. Table or figure (limited to one)
Send all manuscripts by e-mail to immuno@usa.redcross.org
Immunohematology
JOURNAL OF BLOOD GROUP SEROLOGY AND EDUCATION
Instructions to the Authors
Immunohematology
Instructions for Authors
Musser Blood Center
FIRST CLASS
U.S. POSTAGE
PAID
SOUTHERN MD
PERMIT NO. 4144

Colton Chido Rogers

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Colton Chido Rogers

Uploaded by

Copyright:

Available Formats

Journal of Blood Group Serology and Education

Volume 26, Number 1, 2010

You might also like