TRANSFUSION
MEDICINE
REVIEWS
Vol 12, No. 3
Chromosome Location of Genes Encoding Human Blood Groups
Marion E. Reid, Kirk McManus, and Teresa Zelinski
ORE THAN 250 human blood group anti-
M gens have been described. Of these, more
than 200 have been classified into 23 blood group
systems by the International Society of Blood
Transfusion (ISBT) Working Party on Terminology
for Red Cell Surface Antigens. Over the years, the
chromosomal locations of the genes encoding pro-
teins that express or transferases that give rise to the
blood group systems have been determined. Twenty-
one of these genes are located on 12 autosomes,
and two are located on the X chromosome. The
genes encoding all but three blood group systems
(Dombrock, Scianna, and P) have been cloned and
sequenced, and the molecular basis of many indi-
vidual antigens within these systems has been
resolved. The purpose of this review is to update
the reader on the chromosomal location of genes
encoding human blood groups, to provide an
overview of the methods used to assign a gene to a
specific location, to provide information on highly
polymorphic DNA markers in areas surrounding
blood group genes, and to discuss the naming of
blood group genes and their products.
Historically, investigations of blood groups have
played a major role in the development of the
discipline known as human genetics. For example,
the study of ABO blood groups provided proof of
Mendelian inheritance of a normal characteristic in
humans? The first example of autosomal linkage in
humans was reported between L U and S E in 19512
and confirmed in 1958. 3 Through analysis of the
L U : S E linkage relationship, the first proof of
crossing-over between homologous chromosomes
was provided. 3 Establishment of R H : E L 1 linkage 4
in some families in which elliptocytosis occurred
and nonlinkage 5 in others was the first indication
Transfusion Medicine Reviews, Vo112, No 3 (July), 1998: pp 151-161
July 1998
that an inherited disease could have two genetic
backgrounds. 6 After the initial classification of the
autosomes, FYbecame the first gene to be assigned
to a specific chromosome through linkage with an
inherited variation in heterochromatin. 7 Further-
more, X G was the first gene controlling a normal
characteristic to be assigned to the X chromosome
by the study of sex-linked inheritance. 8 Finally, the
first demonstration of a sex difference in recombina-
tion fractions came from analyses of linkage data
involving blood groups, L U : S E 9 and A B O : N P S . io
Just as investigations of blood groups have
enhanced the field of human genetics, the reverse
has also occurred. Technical advances in genetics
and molecular biology have provided blood group
scientists the opportunity to further their discipline.
As a result, information regarding the chromo-
somal location, DNA sequence, molecular varia-
tion, and regulatory elements of genes controlling
blood group expression has been elucidated.
The primary objective of this article is to provide
an update on the chromosomal location of genes
encoding human blood groups. Because consider-
ation of this topic cannot occur without prior
From the Department of lmmunochemistry, New York Blood
Center, New York, NY (M.E.R.); and the Rh Laboratory,
Department of Pediatrics and Child Health and Department of
Human Genetics, Faculty of Medicine, University of Manitoba,
Winnipeg, Canada (K.M., T.Z).
Supported in part by grant no. HL-54459 from the National
Institutes of Health (M.E.R.) and by the Winnipeg Rh Institute
Foundation and the Children's Hospital Foundation (ZZ).
Address reprint requests to Marion E. Reid, PhD, Immune-
chemistry, New York Blood Center, 310 East 67th St, New York,
NY IO021.
Copyright 9 1998 by W.B. Saunders Company
0887- 7963/98/1203-0001 $3.00/0
151
152
discussion of the relevant genetic methodologies,
our review begins by detailing the appropriate
background information.
CHROMOSOMAL BANDING PATTERNS
In 1998, the assignment of individual genes to
specific human chromosomes and their sublocaliza-
tion to discernible bands within the chromosome is
taken somewhat for granted. The precision of gene
assignments/localizations of today is greatly supe-
rior to that in the past, when gene assignments were
restricted to sex chromosomes or some undeter-
mined autosome. The field of human cytogenetics
was changed forever when, in 1956, Tijo and
Levan n established the normal chromosomal
complement of human zygotes as either 46,XX or
46,XY. They reported that each human chromo-
some possessed a unique series of cytogenetic
identification marks that could accurately distin-
guish one from another. These individual chromo-
some banding patterns reflected variation in the
longitudinal structure of different chromatids. De-
pending on the type of dye used, different banding
patterns were obtained. The most common are Q-
and G-bands resulting from the use of quinacrine
and giemsa dyes, respectively. The dark-staining
G-bands are AT-rich regions of highly condensed
DNA commonly referred to as heterochromatin
(the genetically inert region), whereas the light-
staining areas are euchromatin (the genetically
active region).
Guidelines for documenting chromosomal band-
ing patterns have been established by the Interna-
tional Society for Cytogenetic Nomenclature (ISCN)
Committee. 12 Briefly, the centromere is the "cen-
tral" reference site of each chromosome as it
divides the chromosome into two arms. The short
arm (always displayed on the top of a karyogram) is
termed p (petit), and the long arm (always dis-
played on the bottom of a karyogram) is termed q
(queue). Starting at the centromere, successive
bands are numbered in a consecutive fashion. For
example, the bands next to the centromere are
referred to as pl and ql, and the distal ends (those
closest to the telomeres) have higher numbers (eg,
p3 and q3). Further resolution of each band shows
the presence of additional bands termed sub-bands.
These sub-bands are also numbered in a sequential
manner radiating from the centromere. For in-
stance, if there were three sub-bands found within
band pl, they would be labeled p l l , p12, and p13,
REID, McMANUS, AND ZELINSKI
respectively. Sub-band pl 1 is nearest to the centro-
mere, and pl 3 is the most distal. An additional level
of chromosomal banding distinction, that is sub-sub-
bands, is also possible. Numbering occurs as for the
sub-bands. For instance, if p l l had three sub-sub-
bands, they would be labeled from the centromere
to the telomere, with a decimal point distinguishing
them from the sub-bands (eg, pll.1, pll.2, and
pll.3).
Clearly, more detailed chromosomal banding
patterns depend on the level of resolution. Resolv-
ing power can be increased by using more elon-
gated chromosomes. Bands tend to progressively
coalesce to form fewer but thicker bands, as
chromatin condensation proceeds through mitosis.
For instance, a chromosome at metaphase (highly
condensed) would display less resolution, and
therefore fewer bands, than a chromosome at
prometaphase (less highly condensed). In some
high-resolution chromosome banding protocols,
2,000 AT-rich bands can be distinguished, but
typical patterns yield 400, 550, or 850 bands. As a
molecular point of reference, one chromosomal
band from a 400-band karyogram would contain
approximately 7.5Mb DNA.
GENE LOCALIZATION METHODS
Human genes have been localized to individual
chromosomes by a variety of methods, including
somatic cell hybridization, in situ hybridization,
chromosomal anomalies (translocations, deletions,
and monosomies), X-linked inheritance, and link-
age analysis. All of these methods are described
briefly to give the reader an overview of the
techniques involved in localizing not only blood
group genes, but any gene of interest.
Somatic Cell Hybridization
The principle of somatic cell hybridization is
based in the fusion of different somatic cells from
the same or different species. The result is a hybrid
cell line that can be used for gene mapping
purposes. Typically, human fibroblast cells are
combined with mouse tumor cells to produce a
binucleated fusion cell product called a hetero-
karyon. As the heterokaryon continues through
mitosis, dissolution of the nuclear membranes
allows the sister chromatids of each species to
migrate to opposite ends of the cell. On cytoken-
esis, two daughter cells (synkaryons) are produced,
each containing only one nucleus made up of
BLOOD GROUP GENES
chromosomes from both species. Synkaryons are
known to be unstable cells, as most of the human
chromosomes fail to replicate and are lost at
random through successive mitotic events. Any
stable cell lines that do erfierge are therefore
composed of various numbers of different human
chromosomes and constitute the hybrid cell panel.
Some cell lines may only contain one chromosome,
and others may contain several. These hybrid
panels are then used to map hmnan DNA sequences
to a particular chromosome by one of the following
methods: a DNA probe can be labeled and hybrid-
ized directly to the panel, or a polymerase chain
reaction (PCR) assay with primers for a known
DNA sequence is performed and the presence or
absence of amplified product is determined for each
hybrid cell. For somatic cell hybridization to be of
value in gene localization studies, the exact human
chromosome content of each hybrid must be estab-
lished. Even by doing so, somatic cell hybridization
is considered a "gross" mapping method, only
identifying the pertinent chromosome.
In Situ Hybridization
In situ hybridization is a "finer" mapping tech-
nique that resolves the chromosomal position of
human genes through the use of a radiolabeled
probe. Human metaphase chromosomes are air
dried onto a microscope slide before being stained
to reveal their banding pattern. Once stained, they
are photographed. Residual RNA is digested from
the chromosomes with ribonuclease A, leaving the
chromosomal DNA vulnerable to denaturation
agents. After denaturation, a sequence-specific radi-
olabeled probe is added to the fixed chromosomal
spread (the probe will hybridize to its complemen-
tary sequence). Washing removes excess probe,
and a photographic emulsion is placed on top of the
slide. An autoradiograph is made, and Ionce devel-
oped, is placed over the original banding pattern
photograph. The developed silver grains in the
autoradiograph can be juxtaposed with the banding
pattern, thereby identifying the chromosome and
specific band containing the gene of interest. The
radioisotope recommended for in Situ hybridization
is 3H and not 32R because high-emission isotopes
cause the autoradiographic signal to scatter. How-
ever, because of the weak emission characteristics
of 3H, long development times are required, typi-
cally from weeks to months.
Fluorescent in situ hybridization (FISH), a
153
method that evolved from in situ hybridization, is
another tool used in gene mapping studies. Meta-
phase chromosomes are prepared in the same
manner as described in the previous paragraph,
except that the DNA probe used in FISH is labeled
with modified nucleotides. These nucleotides are
covalently linked via a long spacer group to a
reporter molecule (such as biotin or digoxigenin).
On removal of excess probe, the chromosome
preparation is bathed in a solution containing a
fluorescently labeled affinity molecule such as
avidin conjugated to a fluorophore. When exposed
to the proper wavelength, the probe-avidin com-
plex fluoresces. Compared with in situ hybridiza-
tion, FISH has many advantages; it is possible to
obtain results rapidly, it is safer, because radioiso-
topes are not used, numerous DNA clones/seg-
ments can be simultaneously mapped by using
different fluorophores (ie, different colors), and,
depending on resolution, the relative order of
certain genes may be determined in one experi-
ment. However, there are some disadvantages of
FISH. To get signals of sufficient intensity, large
DNA probes ( - 4 0 kb) must be used. Within these
large DNA probes may be regions of repetitive
DNA that may hybridize in different chromosomal
regions, thereby giving multiple signals. To mini-
mize this effect, repetitiVe sites are blocked by
flooding the slide with an excess of unlabeled total
genomic DNA before denaturation. The competi-
tive hybridization that results through the satura-
tion of repetitive elements within the probe allows
visualization of the unique site. Despite this techni-
cal difficulty, FISH has become a popular method
for gene localization studies.
Chromosomal Aberrations
Phenotypic variation may indicate an underlying
chromosome aberration. For example, loss or al-
tered expression of an autosomal dominant trait
may be explained by chromosome translocation. In
this process, breaks occur in two chromosomes
with a portion of one reattaching to the other and
vice versa. Cytogenetic analysis identifies the two
chromosomes involved and dictates that subse-
quent gene localization studies focus on them.
Similarly, chromosomal breakage and segmental
loss coupled to a concurrent phenotypic alteration
suggests that the controlling gene resides within the
lost segment. Finally, reduced expression of an
inherited trait in individuals monosomic for a
154
particular chromosome makes it a candidate for
housing the gene of interest.
X-Linked Inheritance
Typically X-linked inheritance andconsequently
gene assignments are resolved by scrutiny of
individual families segregating for the genetic trait
in question. Characteristics of X-linked inheritance
may include, but are not restricted to, the follow-
ing: Either sex may express the trait, but there are
usually more females than males, because daugh-
ters may inherit the trait from either parent, whereas
sons can only inherit the trait from their mothers.
Because of random X inactivation (lyonization),
females usually express the trait mildly or variably
as compared with males. Daughters or sons of a
woman possessing an X-linked trait have a 50%
Chance of inheriting it. Daughters of a man possess-
ing an X-linked trait have a 100% chance of
inheriting it, whereas sons of the same man will
never inherit the trait.
Linkage Analysis
The chromosomal location of many human genes
has been established and/or confirmed through
linkage analysis. The aim of linkage studies is to
evaluate the segregation of two heritable traits (eg,
genes, restriction fragment length polymorphisms
[RFLPs], variable number of tandem repeat poly-
morphism [VNTRs], nucleotide repeats). Genes are
said to be linked when they are on the same
chromosome; however, establishing linkage de-
pends on the proximity of the segregating units. For
instance, genes located on the same chromosome
but at opposite ends may be no more likely to be
inherited together than genes located on different
chromosomes. When two genes reside far apart,
there is an equal chance that they will segregate
either independently because of cross-over events
between them or as alsingle unit. However, for two
genes residing in close proximity, crossing-over is
less likely, and hence these traits should be inher-
ited together. Linkage is inferred when there is a
significant deviation from the expectation that two
genes would segregate independently 50% of the
time. In other words, the two genes appear to
segregate together more frequently than the ex-
pected 50% of the time. For a family to be
informative in a linkage study, at least one parent
must be heterozygous for the two markers in
REID, McMANUS, AND ZELINSKI
question. Informative families are classified as
either phase-known or phase-unknown.
In phase-known families, the allelic alignment is
followed through three or more successive genera-
tions. Direct counts of nonrecombinants (NRs), that is,
maintained grandparental alignments, and recombi-
nants (Rs), that is, disrupted alignments caused by
crossing-over between homologous chromosomes
in meioses, can be determined. If the total number
of Rs:NRs from all phase-known families in the
analysis differs significantly from a 1:1 ratio,"finkage
is established. The genetic distance between the
markers is estimated from the recombination frac-
tion (0) and is calculated by the number of recombi-
nants divided by the number of recombinants added
to the number of nonrecombinants, as follows:
R
R +NR
Because few phase-known families are avail-
able, most linkage information is accumulated
through investigations of two-generation families
(phase-unknown). Once again, allelic alignments
(so-called z counts) are determined in the children
of the informative parent. But, without knowing the
grandparental alignment, it is impossible to deter-
mine whether the z count represents NRs or Rs.
Therefore, z counts for all informative families are
subjected to a statistical analysis that gives an
estimate of the likelihood of linkage. This estimate
is expressed as "lods," the logarithm of the odds in
favor of or against linkage, for the pair of traits in
question. The lods for all phase-unknown families
are summed, and if the value exceeds 3.00 (repre-
senting odds of 1,000:1 in favor of linkage for the
pair of markers), linkage is inferred. A lod of - 2 . 0 0
represents odds of 100:1 against linkage and is used
to exclude linkage between the two genes. Inconclu-
sive lods (between - 2 . 0 0 and 3.00) dictate that
additional informative families be analyzed.
CHROMOSOMAL LOCATIONS
OF BLOOD GROUP GENES
The chromosomal locations of 23 blood group
system genes 3 are depicted in Figure 1. With one
exception (the DO assignment to chromosome 1214
has not been confirmed), each location has been
refined through the efforts of many blood group
laboratories using the methods already described.
Specific details of the "evolving" chromosomal
assignments for individual blood group system
 BLOOD GROUP GENES 155

Chrom~ome 1 2 4 6

I 01~$49
DIS77
Sho~
Arm
(P)

21.3 CHARG I D6S1(~

~ n ~

APOA2
MUCI D2S~D2S93
Long
(q)
28
C48PA~ FABP2
MCP 31:,

7 9 11 12 17

CO ITCRGV1
~AB 13 IN I Dl1S417
DllS907 11;". ,PR81 IVV~=
I EGFR

21 D/ D47S40
D17S41
YT I COLIA2
/ CFTR D7S23
KEL 342I O ASS
ABL1

18 19 22 X
223 PABX
XG DXSlS
133 ~KIJE.LW D19S177 XK I CYBB
132 ! D19S221 9 OTC

i~z IBCR1
D18S34
D18S36 132 I APOC2 P1 j D22S5~
133 LU 019S178 1

Fig 1. Chromosome diagram depicting the localization of various blood group genes and DNA markers, Individual chromosomes
are numbered. The ong and short arms of each chromosome are separated by the horizontal line (centromere). Blood group genes
(italicized) not localized to a specific chromosomal band are encased in brackets, indicating the boundaries of their assmgnment.
Vertical lines designate positions of the polymorphic DNA markers listed in Table 1.
 156
genes can be found elsewhere. 15,16Also shown in
Figure 1 are the positions for three genes (RD,
MER2, and OK) controlling the expression of the
low-incidence red cell antigen Rd and the high-
incidence red cell antigens MER2 and Ok a. To date,
the assignments of MER217 to chromosome 11 and
OK to chromosome 1918,19 have not been con-
firmed.
Also depicted in Figure 1 and listed in Table 1
are a series of polymorphic DNA markers located
within the chromosomal regions of interest. Most
of these DNA markers have been designated as
reference markers (underlined in Table 1) by the
Human Gene Mapping (HGM) DNA Committee. 2~
Reference markers are intended to be used as a
defined set of chromosomal "landmarks" for ge-
netic linkage studies. It is hoped that provision of a
well-characterized standard set of highly informa-
tive markers (heterozygosity values of 0.75 or
above), evenly spaced throughout the genome, will
facilitate exchange of genetic mapping information
between investigators in different fields.
For the purpose of this article, a hierarchical
process for the selection of DNA markers was used.
Preference was given to reference DNA markers in
which linkage relationships between them and the
blood group gene have been established (eg, YT:
COLIA2, 21 DI:D17S41, 22 and LU:APOC223). The
selection continued with reference markers that
reside in the region of interest, because they are
linked to other genes or anonymous markers in the
area, but for which linkage relationships between
them and the blood group gene have not yet been
established (eg, CO:EGFR, DO:PRB1, and LW:
D19S20). If no reference markers were found
within the blood group gene location, a nonref-
erence marker (eg, CROM, KN:MCP,, and CO:
TCRGV1) was selected based on the following
criteria: it is polymorphic, it has a high heterozygos-
ity value, it resides within the chromosomal region
of the blood group gene, and it has a well-
established detection method (ie, size of PCR
amplification productl or different restriction endo-
nuclease sites).
For each of the blood group genes, two polymor-
phic DNA markers are listed in Table 1. These
markers can be used in genetic investigations of
human blood groups in two major ways. First,
linkage studies can be conducted in which families
segregating for the blood group gene of interest are
REID, McMANUS, AND ZELINSKI
analyzed for the DNA markers listed and for other
DNA markers (ascertainment discussed in the next
section) within the region. The results of this type
of investigation would provide valuable informa-
tion on the genetic relationships between markers
(multi-point mapping) and also help determine why
certain chromosomal regions, including some that
contain blood group genes, appear to be more
meiotically active (eg, evidence of excess recombi-
nation within lp36-p2224,25 and 19p13A-q13.123
has been observed). Because the current genetic
maps are based largely on relationships between
nonexpressed genes (ie, anonymous DNA seg-
ments), it seems necessary to expand this informa-
tion by the inclusion of classical genetic markers
such as blood groups. A detailed analysis of the
pertinent chromosomal regions would allow for
comparison of genetic relationships between nonex-
pressed and expressed markers. Documentation of
differences could contribute to a further understand-
ing of genome organization. Second, DNA markers
can be used as substitutes for blood group genes in
investigations of either previously or "newly"
described red cell specificities. For example, many .
of the antigens within the ISBT's low- and high-
incidence antigen series are not excluded from the
Diego blood group system. Among this group,
potential Diego system candidates have been previ-
ously reviewed. 26 For the remainder, exclusion
from the Diego blood group system seems unlikely
because of a lack of informativeness (DI is typi-
cally polymorphic in people of Mongolian de-
scent). This difficulty may be overcome by substi-
tuting D17S40 and/or D17S4I for DI. Heter-
ozygosities of 0.4200 for D17S40 and 0.3750 for
D17S41 do not ensure that a given family will
segregate for either, but the chances are dramati-
cally greater than for DI. The same principle can be
applied to studies of any blood group in which the
segregating family is not informative. By conduct-
ing investigations of this type, the amount of
genetic information relative to individual specifici-
ties should be substantially increased, atloTging for
the possible creation of new blood group systems
or inclusion of an established or "new" specificity
within an existing blood group system. The inter-
ested reader can find further examples of the use of
blood group substitute markers in a previous re-
view. 27
BLOOD GROUP GENES 157

Table 1. Polymorphic DNA Markers Surrounding Blood Group Genes

Blood Group Substitute Marker Location VariationType Maximum Heterozygosity
Rh D1S49 1p36-p34 VNTR .7400
DIS77 1p36-p34 VNTR .7090
Scianna DIS49 1p36-p34 VNTR .7400
DIS77 1p36-p34 VNTR .7090
Duffy APOA2 1q21-q23 DINUC .6984
MUC1 1q21-q23 VNTR .9000
Cromer C4BPA/C4BPB lq32 DINUC .7479
MCP lq32 RFLP .4990
Knops C4BPA/C4BPB lq32 DINUC .7479
MCP lq32 RFLP .4990
Gerbich D2S95 2q14-q21 DINUC .8500
D2S93 2q21-q23 DINUC .8300
MNS FABP2 4q28-q31 TETNUC .8030
TRINUC .7980
Chido/Rodgers D6S109 6p22,3-p22.2 DINUC .7890
D6S105 6p22.1-p21.3 DINUC .7870
Colton TCRGV1 7p15-p14 RFLP .6480
EGFR 7p12 DINUC .7500
Yt COL1A2 7q21.3-q22.1 TRINUC .7401
DINUC .6620
Kell CFTR 7q31.3 DiNUC .8850
D7S23 7q31-q32 RFLP .5100
ABO ASS 7q34.1 DINUC .8460
ABL1 7q34.1 DIN UC .6800
Indian D11S907 1lp13 DINUC .7353
D11S417 1lp13 iNS .7358
Dombrock VWF 12p13.3~p13.2 TETNUC .7990
PRB1 12p13.2 INS .6532
Diego D17S40 17q 12-q24 RFLP .4200
D17S41 17q 12-q24 RFLP .3750
Kidd D18S34 18q12,2-q12.3 DINUC .8102
D18S36 18q12,2-q12.3 DINUC .7500
Lewis D19S20 19p13.3 VNTR .8300
D19S177 19p13.3 DINUC .7940
LW D19S20 19p13,3 VNTR .8300
D19S177 19p13.3 DINUC .7940
Lutheran APOC2 19q 13.2 DINUC .8520
D19S178 19q13.2 DINUC .7560
Hh APOC2 19q13.2 DINUC .8520
D19S178 19q13.2 DINUC .7560
P BCR1 22ql I TRINUC .5050
D22S'55 22q 13 VNTR .4970
Xg PABX Xp22.3 RFLP .8200
DXS16 Xp22.2 DIN UC .8734
Kx CYBB Xp21.1 DINUC ,7608
OTC Xp21.1 RFLP .4350

Radin D1S49 1p36-p34 VNTR .7400

D1S77 1p36-p34 VNTR .7090
MER2 D11S922 1lp15.5 DINUC .9353
HRAS 1lp15.5 VNTR .8400
OK D19S221 19p13.2 DINUC .8710
EPOR 19p13.2 DINUC .8718
NOTE. Reference markers are underlined. Other markers were selected based on high heterozygosity values. D represents DNA,
followed by chromosome number, S represents segment, and the final number is the chronological numbering of the segment.
Abbreviations: VNTR, variable number of tandem repeat polymorphism; DINUC, dinucleotide repeat polymorphism; TRINUC,
trinucleotide repeat polymorphism; TETNUC, tetranucleotide repeat polymorphism; RFLP, restriction fragment length polymor-
phism; INS, insertion polymorphism.
158
SEARCHING THE GENOME DATABASE
If the DNA markers listed in Table 1 are not
segregating in the families of interest, other poten-
tial substitute markers can be found in the Genome
Database (GDB). As of December 1997, there were
more than 18,000 polymorphic DNA markers (D
segments or genes) characterized in GDB. Two
major categories of marker, namely RFLPs and
microsatellites, are described. RFLPs are variations
in nucleotide sequence that create or abolish a DNA
restriction site. The resultant fragments (alleles) are
of different lengths and are typically detected on
agarose gels. Average heterozygosities for RFLPs
have been calculated as 0.40. Microsatellite mark-
ers are nucleotide repeat polymorphisms that consti-
tute approximately two thirds of all markers within
the database. They have been subcategorized ac-
cording to the number of units in the repeat (ie,
mononucleotide repeats [MONUC], dinucleotide
repeats [DINUC], trinucleotide repeats [TRINUC],
tetranucleotide repeats [TETNUC], pentanucleo-
tide repeats [PENUC], and VNTR). The most
recent compilation 28 of microsatellite markers is
set out in Table 2. Approximately 92.5% of all
microsatellites are found within two subcategories
(DINUC and TETNUC), representing approxi-
mately 60% of the total number of polymorphic
markers found in GDB. Average heterozygosities
for microsatellites range from 0.59 (MONUC) to
0.71 (DINUC). Heterozygosities for the remaining
subcategories are in the range of 0.63 to 0.64. There
appears to be an inverse relationship between the
average number of alleles and the number of units
in the repeat. MONUC and DINUC average seven
alleles each, whereas PENUC average only four.
Table 2. Categories of Microsatellite Markers Characterized
in the Genome Database
MONUC DINUC TRINUC TETNUC PENUC VNTR
No. in GDB 12 7378 860 3789 27
Average cal-
culated
heterozy-
gosity 0.59 0.71 0.64 0.63 0.64
Average no.
of alleles 7 7 5 5 4 6
Abbreviations: GDB, genome database; MONUC, mono-
nucleotide repeat polymorphism; DINUC, dinucleotide repeat
polymorphism; TRINUC, trinucleotide repeat polymorphism;
TETNUC, tetranucleotide repeat potymorphism; PENUC, penta-
nucleotide repeat polymorphism; VNTR, variable number of
tandem repeat polymorphisms.
REID, McMANUS, AND ZELINSKI
Blood group investigators requiring additional
DNA-based polymorphic markers for their studies
can access the information on-line using a Java-
enabled browser that is 3.0 or above (eg, Microsoft
Internet Explorer 3.0+ or Netscape Navigator
3.0+). Steps to searching GDB for additional
markers are as follows:
1. Open the browser and go to the Genome
Database at "http://gdbwww.gdb.org/"
2. Once in the database, click on the "Other
Search Option" button.
3. Once the next screen has fully loaded, click
on the "Find Region of Interest" button under
the "Map" section of the search options.
4. You will be given the option of selecting a
portion of a particular chromosome or select-
ing an entire chromosome. For example, if
you wanted to find a marker that is kilown to
reside between lp36 and lp34, you would
select "FROM: lp36 TO: lp34." Another
option is the type of map to be retrieved.
Under the "Limit maps to those containing
above markers," you may select either "re-
trieve only maps containing markers" or
"retrieve all maps of region." Finally, you"
also may select the type of map displayed by
clicking on the desired map type (eg, cytoge-
netic, linkage, radiation hybrid, contig or
integrated) under the "Limit maps to the
following types" section. Once all of these
options have been entered according to your
criteria, click on the "submit" button at the
bottom of the page.
5. To view the desired map(s), simply click on
the map(s) of interest and then on the "dis-
play maps" button.
6. Once selected, the browser will open a Java
applet. It will take a few minutes for the
applet to download because the file is approxi-
mately 370 kb. You will see a new window,
414
entitled "GDB Mapview 2.3."
7. Once the map(s) has fully loaded, including
all the elements (genes, markers, etc.), select
0.63 an element of interest by clicking, on--it. Once
selected, a red box will appear around the
element. For additional information on the
element, double click on the highlighted
element. For example, double clicking on
marker D 1$49 will open an additional browser
window entitled "Clone CRI-L1046." Fur-
ther infocmation on D1S49 (eg, other names,
BLOOD GROUP GENES 159

GDB accession number, cytogenetic location,
polymorphisms) is detailed. A simple table
characterizing the polymorphism is given.
More information also can be found by click-
ing on the polymorphism Within the table. For
instance, in the case of D1S49, if the "D1S49
Variable Number of Tandem Repeats" poly-
morphism is selected and clicked, a new
screen will appear. This screen is entitled
"POLYMORPHISM D1S49 Variable Num-
ber of Tandem Repeats." Here, additional
information on the type of variation, the
polymorphism detection method(s), allele sets,
maximum heterozygosity, etc. can be found.
Although the steps just outlined will facilitate
selection of additional DN A markers, it should be
kept in mind that databases are dynamic entities.
From time to time the query forms may change.
Therefore, the reader is advised to use these steps
as a guideline only.
BLOOD GROUP GENES
AND GENE PRODUCTS
Table 3 lists the chromosomal locations of 23
blood group system genes and the genes coding for
three other antigens, as well as defining the prod-
ucts (ie, proteins that express or transferases that
give rise to the blood group specificities) of these
genes. In addition~ the formal gene designations of
both the ISBT 13 and the Human Gene Mapping
(HGM) 2~nomenclature committees are listed. There
has been some confusion as to when the ISBT or
HGM gene designations should be used by investi-
gators. If the product of the gene has been studied
by serological or immunochemical methods, the
correct gene designation is based on the ISBT

Table 3. Products of Genes Controlling Blood Group Expression

ISBT Blood GroupSystem GeneName
Chromosomal Antigens of ISBTLow-
Location or High-IncidenceSeries ISBT HGM GeneProduct
1p36.1-p34.3 Rh (004) RH RHD Acylated RhD protein
RHCE Acylated RhCE protein
1p35-p32 Scianna (013) SC SC Sc Glycoprotein
1q22-q23 Duffy (008) FY DARC Fy Glycoprotein
lq32 Cromer (021) CROM DAF CD55 (DAF)
lq32 Knops (022) KN CR1 CD35 (CRI)
2q14-q21 Gerbich (020) GE GYPC GPC & GPD
4q28.2-q31.1 MNS (002) MNS GYPA, GYPB GPA, GPB
G YPE GPE
6p21.3 Chido/Rodgers (017) CH/RG C4A C4A
C4B C4B
7p14 Colton (015) CO AQP1 Channel-forming integral protein
7q22 Yt (011) YT ACHE Acetylcholinesterase
7q33 Ketl (006) KEL KEL Kell glycoprotein
9q34.2 ABO (001) ABO ABO A - c~-3-N-acetyl-g-galactosaminyltransfer-
ase
B = c~-3-D-galactosyltransferase
11 p13 Indian (023) IN CD44 CD44
12p13.2-p12.1 Dombrock (014) DO DO Do glycoprotein
17q21-q22 Diego (010) DI SLC4A 1 Anion exchanger 1 (band 3)
18ql 1-q 12 Kidd (009) JK SLCI4A 1 Urea transporter
19p13.3 Lewis (007) LE rUT3 c~-3/4-L-fucosyltransferase
19p13.3 Landsteiner-Wiener (016) LW LW LW glycoprotein
19q13.2 Lutheran (005) LU LU Lutheran glycoprotein and B-CAM
19q13.3 Hh (018) H rUT1 c~-2-L-fucosyltransferase
22q 11.2-qter P (003) PI P1 Galactosyltransferase
Xp21.1 Kx (019) XK XK Kx glycoprotein
Xp22.3 Xg (012) XG XG Xg a glycoprotein

1p35-p32 Radin (700015) RD RD Rd glycoprotein

1lp15.5 MER2 (901011) MER2 MER2 Not defined
19pter-p13.2 OK (901006) OK CD147 CD147
Abbreviations: ISBT, International Society of Blood Transfusion; HGM, human gene mapping.
160
symbol for a blood group system and is written in
italics (eg, RH, KEL, FY). Similarly, for serological/
immunochemical investigations involving a spe-
cific antigen, tile system symbol and antigen num-
ber are given in italics (eg, RH 1, KEL 2, FY 1). In
general, when investigations of the gene or gene
product have been conducted by molecular meth-
ods, the HGM gene name should b e used. For
example, the symbol for an allele within a particu-
lar blood group system is designated by the gene
name, followed by an asterisk and letter or number
in italics (eg, RH*D, KEL*2, DARC*A).
Although it may be difficult to remember the
rules for using the two gene nomenclatures, there
seems to be some merit in maintaining each system.
For serologists, using the ISBT designations 13 is a
reminder of the evolution of blood group science
and the well-characterized erythrocyte specificities
that are still determined today. For molecular
biologists, the HGM designations 2~ provide clues
to the character or function of a gene product. For
example, the HGM gene designation for the Diego
blood group system is SLC4A1 (solute carrier
family 4, anion exchanger, member 1), and that for
the Kidd blood group system being SLC14A1
(solute carrier family 14, urea transporter, member
1). The gene names reflect the relationship between
their products (ie, both are transmembrane mol-
ecules involved in solute exchange). If, for ex-
ample, a new gene controlling the expression of a
similar product on the surface of erythrocytes was
described, it is likely to be designated as SLC
followed by the appropriate family and member
number. It is worth mentioning that as of July 1997,
6,651 human genes have been named by HGM. 29
Because the list of gene designations is growing at
a rapid pace, investigators should contact the HGM
Nomenclature Committee before naming any gene
(the names and contact information for the appropri-
ate HGM committee members are provided by
White et al.2~ Finally, it is important for investiga-
tors to realize that HGM gene designations change
as more functional information becomes available.
For example, although the ISBT gene name for the
Diego blood group system (DI) has not changed
over the years, the HGM name has, from DI to
EPB3 (erythrocyte surface protein, band 3) to AE1
(anion exchanger 1), to the current SLC4A1.
Antigens of three blood group systems (Chido/
Rodgers, Rh, and MNS) listed in Table 3 are
encoded by two homologous genes that are adja-
cent on the same chromosome. Because of this, the
REID, McMANUS, AND ZELINSKI
genes and gene products of these systems can
include hybrids. 3~ For the Chido/Rodgers blood
group system, the two common gene names and the
encoded products are listed. In the Rh blood group
system, the ISBT gene name of RH is used despite
the fact that two gene products carrying the D and
Cc/Ee antigens are known. The HGM gene names
denote the two distinct genes (RHD and RHCE).
Similarily, for the MNS system, the ISBT gene
name of MNS is used for both the genes and for the
two products (GPA and GPB): GPA carrieg the M/N
polymorphism, and GPB carries the S/s polymor-
phism. The HGM system names both genes (eg,
GYPA encodes GPA and GYPB encodes GPB). A
third gene in this family (GYPE) has not been
shown conclusively to encode its product (GPE) in
the erythrocyte membrane. 32
The products of each gene listed in Table 3 refer
to all polymorphic forms, the number of which
ranges from one antigen (P blood group system) to
nearly 50 antigens (Rh blood group system). The
exception is in the ABO blood group system, where
the two alleles of ABO encode different sugar
transferases.
SUMMARY
The chromosomal locations of genes controlling
the expression of some 200 antigens constituting
the 23 established human blood group systems
have been reviewed. Twenty-one of the these genes
are located on 12 autosomes, and two are located
on the X chromosome. Refihed chromosomal posi-
tions, to a single cytogenetically distinguishable
band, have been established for 13 of the 23 genes.
For the remainder, continued investigation will
achieve the same result. The genes (RD, MER2, and
OK) controlling the expression of one low-
incidence and two high-incidence erythrocyte anti-
gens have also been presented. Of these, OK is the
most likely candidate for blood group system
status, because its chromosomal location distin-
guishes it from all established system genes except
LE and LW, and, the product of the OK gene is
different from those of LE and LW (Table 3): This
issue will be considered at the next meeting (sched-
uled for July 1998) of the ISBT Working Party.
Alternatively, RD and MER2 are not good candi-
dates for blood group system status because RD
and MER2 reside in chromosomal regions contain-
ing genes for other blood group systems. In addi-
tion, the products of RD and SC have similar
biochemical characteristics, 33 and the product of
BLOOD GROUP GENES
MER2 has not yet been defined (Table 3). The
challenge remaining for blood group scientists is
characterization of genes that control expression of
the approximately 50 other known erythrocyte
antigens. Most of these are members of the ISBT's
700 (low-incidence) or 901 (high-incidence) se-
ries. 13 Because the current genetic information for
each of these antigens (attained by serologic inves-
tigation) varies considerably, future studies will
REFERENCES
1. Epstein AA, Ottenberg R: Simple method of performing
serum reactions. Proc N Y Path Soc 8:117-123, 1908
2. Mohr J: A search for linkage between the Lutheran blood
group and other hereditary characters. Acta Pathol Microbiol
Scand 28:80-96, 1951
3. Sanger R, Race RR: The Lutheran-secretor linkage in
man: Support for Mohr's findings. Heredity 12:513-520, 1958
4. Goodall HB, Hendry DWW, LaMer SD, et al: Data on
linkage in man: Elliptocytosis and blood groups. Ann Eugen
17:272-275, 1953
5. Chalmers JNM, Lawler SD: Data on linkage in man:
Elliptocytosis and blood groups. Ann Eugen 17:267-271, 1953
6. Morton NE: The detection and estimation of linkage
between the genes for elliptocytosis and the Rh blood type. Am J
Hum Genet 8:80-96, 1956
7. Donahue RR Bias W, Renwick JH, et al: Probable
assignment of the Duffy blood group locus to chromosome 1 in
man. Proc Natl Acad Sci U S A 61:949-955, 1968
8. Mann JD, Cahan A, Gelb AG, et al: A sex-linked blood
group. Lancet 1:8-10, 1962
9. Cook PJL: The Lutheran-secretor recombination fraction
in man: A possible sex-difference. Ann Hum Genet 28:393-401,
1965
10. Renwick JH, Schulze J: Male and female recombination
fractions for the nail-patella: ABO linkage in man. Ann Hum
Genet 28:379-392, 1965
11. Tijo JH, Levan A: The chromosome number in man.
Hereditas 42:1-6, 1956
12. ISCN: An international system for human cytogenetic
nomenclature. Cytogenet Cell Genet 21:309-404, 1978
13. Daniels GL, Anstee DJ, Cartron J-R et al: Blood group
terminology 1995. From the ISBT working party on terminol-
ogy for red cell surface antigens. Vox Sang 69:265-279, 1995
14. Eiberg H, Mohr J: Dombrock blood group (DO): Assign-
ment to chromosome 12p. Hum Genet 98:518-5211 1996
15. Lewis M, Zelinski T: Linkage relationships and gene
mapping of human blood group loci, in Cartron J-R Rouger P
(eds): Molecular Basis Of Major Human Blood Group Antigens.
New York, NY, Plenum Press, 1995, pp 445-475
16. Daniels G: Blood group gene mapping, in Daniels G (ed):
Human Blood Groups. Oxford, Blackwell Science, 1995, pp
693-710
17. Palmer DK, Miller YE, Geyer D, et al: MER2, a new
human red cell polymorphism controlled by a locus on chromo-
some 11. Cytogenet Cell Genet 40:720, 1985
18. Williams BR Pym B, Tippett P, et al: Another red cell
surface antigen Ok a is encoded by a gene on chromosome 19.
Cytogenet Cell Genet 46:717, 1987
19. Williams BR Daniels GL, Pym B, et al: Biochemical and
161
have to rely on "tools" from related disciplines to
provide the additional information. Use of re-
sources such as molecular biological protocols and
GBD should facilitate the effort.
ACKNOWLEDGMENTS
The authors thank Gail Coghlan for helpful advice and
constructive criticism during the preparation of this manuscript
and Bob Rather for preparing the manuscript.
genetic analysis of the OK a blood group antigen. Immunogenet-
ics 27:322-329, 1988
20. White JA, McAlpine PJ, Antonarakis S, et al: Guidelines
for human gene nomenclature (1997). Genomics 45:468-471,
1997
21. Zelinski T, White L, Coghlan G, et al: Assignment of the
YT blood group locus to chromosome 7q. Genomics 1l:165-
167, 1991
22. Zefinski T, Coghlan G, White L, et al: The Diego blood
group locus is located on chromosome 17q. Genomics 17:665-
666, 1993
23. Lewis M, Kaita H, Coghlan G, et al: The chromosome 19
linkage group LDLR, C3, LW, APOC2, LU, SE in man. Ann
Hum Genet 52:137-144, 1988
24. Lewis M, Kaita H, Chown B, et al: Relative positions of
chromosome l loci Fy, PGM1, Sc, UMPK, Rh, PGD and ENO1
in man. Can J Genet Cytol 19:695-709, 1977
25. Lewis M, Kaita H, Philipps S, et al: The position of the
Radin blood group locus in relation to other chromosome i loci.
Ann Hum Genet 44:179-184, 1980
26. Zelinski T: Erythrocyte band 3 antigens and the Diego
blood group system. Transfus Med Rev 12:36-45, 1998
27. Zelinski T: The use of DNA restriction fragment length
polymorphisms in conjunction with blood group serology.
Transfusion 31:762-770, 1991
28. Bowcock AM, Chipperfield MA, Ceverha P, et al: Report
of the DNA Committee, in Cuticchia A J, Chipperfield MA,
Foster PA (eds): Human Gene Mapping 1995: A Compendium.
Baltimore, MD, Johns Hopkins University Press, 1996, p 1455
29. Blake JA, Davisson MT, Eppig JT, et al: A report on the
International Nomenclature Workshop held May 1997 at the
Jackson Laboratory, Bar Harbor, Maine, U.S.A. Genomics
45:464-468, 1997
30. Huang C-H: Molecular insights into the Rh protein
family and associated antigens. Current Opinion in Hematology
4:94-103, 1997
31. Daniels G: Blood group antigens as markers of comple-
ment and complement regulatory molecules, in Cartron J-E
Rouger P (eds): Molecular Basis of Human Blood Group
Antigens. New York, NY, Plenum Press, 1995, pp 397-419
32. Huang C-H, Blumenfeld OO: MNSs blood groups and
major glycophorins: Molecular basis for allelic variation, in
Cartron J-P, Rouger P (eds): Molecular Basis of Major Human
Blood Group Antigens. New York, NY, Plenum Press, 1995, pp
153-183
33. Spring FA: Characterization ofblood-group-active eryth-
rocyte membrane glycoproteins with human antisera. Transfus
Med. 3:167-178, 1993