J.M. Ellis
S.J. Mack
R.F.G. Leke
I. Quakyi
A.H. Johnson
C.K. Hurley
Key words:
Africa; Cameroon; class I; sequence-based typing
(SBT); human leukocyte antigen (HLA); sequencespecific oligonucleotide probe (SSOP) typing
Acknowledgments:
The authors would like to express their thanks to
Henry Erlich for advice regarding the population
analysis, Noriko Steiner for designing the HLA-A
locus AIN-A, T, and G primers, and Christina
Ginsberg for extracting DNA from the CS series
samples. Noriko Steiner and Carol Kosman
sequenced the alleles from unrelated patient and
registry volunteers possessing B*1403, *4703,
and *4016. This research is supported by
funding from the Office of Naval Research
N00014–94–1–0049 to the C.W. Bill Young
Marrow Donor Recruitment and Research
Program (C.K.H.) and NIAID NIH UO1-AI-35839
(I.A.Q., R.F.G.L., A.H.J.). The views expressed in
this article are those of the authors and do not
reflect the official policy of the Department of
the Navy, the Department of Defense, or the
U.S. Government.
Diversity is demonstrated in class I HLA-A
and HLA-B alleles in Cameroon, Africa:
description of HLA-A*03012, *2612,
*3006 and HLA-B*1403, *4016, *4703
Abstract: To examine the genetic diversity in west Africa, class I HLA-A
and HLA-B alleles of 92 unrelated individuals from two areas in the Cameroon, the capital Yaoundé and the village of Etoa, were identified by
direct automated DNA sequencing of exons 2 and 3 of the HLA-B locus
alleles and sequence-specific oligonucleotide probe (SSOP) and/or sequencing of the HLA-A locus alleles. HLA-A*2301 (18.7%), A*2902 (10.4%),
B*5301 (10.9%), and B*5802 (10.9%) were the most frequently detected
alleles, present in at least 10% of the population. A total of 30 HLA-A locus
and 33 HLA-B locus alleles, including six novel alleles, were detected. The
novel alleles were HLA-A*03012, A*2612, A*3006 and HLA-B*1403, B*4016,
and B*4703. HLA-B*4703 contains a novel amino acid sequence that is a
combination of the first 5 amino acids of the Bw6 epitope and the last 2
residues of the Bw4 epitope. The addition of 6 alleles to the ever-expanding
number of known class I HLA alleles supports our hypothesis that extensive
genetic diversity, including previously undescribed alleles, would be observed in this African population. In the Yaoundé population, the allele frequency distribution at the HLA-A locus is consistent with distributions
indicative of balancing selection. Extensive HLA-A-B haplotypes were observed in this population suggesting that only a fraction of the Cameroon
HLA-A-B haplotype diversity has been observed.
Modern-day Homo sapiens likely arose from a population of individuals who carried the ancestral forms of the major human leukocyte
antigen (HLA) allele families. The polymorphism of the HLA system could have further developed through the evolutionary mechanisms of intraallelic and interallelic conversion, reciprocal recombination, and point mutation, under the influence of positive selective
pressure for alleles that efficiently bind and present pathogenic peptides to the immune system (1). Because it has been postulated that
human life arose in Africa (2), the HLA diversity of modern African
Received 16 July 1999, revised, resubmitted,
accepted for publication 31 July 2000
Copyright c Munksgaard 2000
Tissue Antigens . ISSN 0001-2815
Tissue Antigens 2000: 56: 291–302
Printed in Denmark . All rights reserved
Authors’ affiliations:
J.M. Ellis1,
S.J. Mack2,3,
R.F.G. Leke4,5,
I. Quakyi4,
A.H. Johnson6,
C.K. Hurley1
1
Department of Microbiology
and Immunology,
Georgetown University
Medical Center,
Washington, DC, USA,
2
Department of Human
Genetics, Roche Molecular
Systems, Alameda,
California, USA,
3
Children’s Hospital Oakland
Research Institute,
Oakland, California, USA,
4
Department of Biology,
Georgetown University,
Washington, DC, USA,
5
Faculty of Medicine and
Biomedical Sciences,
University of Yaoundé,
Yaoundé, Cameroon,
6
Department of Pediatrics,
Georgetown University
Medical Center, Washington,
DC, USA
Correspondence to:
Carolyn Katovich Hurley,
Ph.D.
E404 Research Building
Georgetown University
Medical Center
3970 Reservoir Rd NW
Washington DC 20007
USA
Tel: π1 202 687 2157
Fax: π1 202 687 6440
populations may best reflect the HLA diversity of the human population prior to the diversification of our species into different racial
and ethnic groups. Significant genetic diversity has been observed
among different African ethnic groups (3, 4), and Africans may dis-
291
Ellis et al : Class I alleles in Cameroon, Africa
play a genetic diversity which is overall greater than that observed
speaks 24 major languages and countless dialects (5). Due to its
for other racial and ethnic groups.
geography and history, Cameroon provides a prime location to
Although Cameroon is located in west Africa, its vivid and di-
study the diversity of class I HLA alleles in Africa.
verse history has been shaped by numerous waves of human migration, making Cameroon a cross-section of all Africa. Situated on
the southern border of Lake Chad, Cameroon lies midway along the
Sahel Corridor (Fig. 1), a major route along the southern border of
Material and methods
the Sahara Desert. Since approximately 2750 B.C., different ethnic
groups – in particular, the nomadic Fulani of modern Senegal and
Whole blood was obtained from 92 unrelated individuals native to
the east African Shuwa Arabs – have used the Sahel Corridor for
Cameroon, primarily living in the capital city of Yaoundé (DT series
travel between east and west coasts of Africa. In approximately 1
samples, n⫽41) and the village of Etoa (CS series samples, n⫽51).
A.D., the so-called outpouring of the Negro tribes was initiated, in
The 41 samples from Yaoundé were derived from unrelated women
which the Zande and the Bantu tribes migrated across Cameroon.
who gave birth at Central Maternity Hospital. The remaining 51
This Bantu expansion resulted in the southern movement of the
samples were selected as unrelated from 200 samples collected dur-
San and the Pygmy peoples, and continued to shape African history
ing a census of the families living in the village of Etoa. Serologic
for nearly a millenium. In the late 16th century, Portuguese ex-
typing was performed on some of the CS series samples (data not
plorers ‘‘discovered’’ Africa, initiating the arrival of European mer-
shown).
chants and missionaries. Over the course of the 18th and 19th cen-
In the case of the HLA-A locus, exons 2 and 3 and intron 2
turies, so-called Islamic holy wars were mounted by the northern
were amplified with the primers 5AIN1–36 and 3AIN3–62 and high-
Fulani, resulting in migration, intermingling and adsorption of dif-
resolution sequence-specific oligonucleotide probe (SSOP) typing
ferent groups in northern Cameroon. In the 19th and 20th centuries,
was performed as previously described (6, 7) using the probes listed
Cameroon was colonized by Germany and France. Cameroon be-
in Table 1. In instances where the HLA types could not be resolved
came independent from France in 1960 (4), and the modern Ca-
to allele level, direct automated DNA sequencing was performed.
meroonian population represents more than 200 ethnic groups, and
By designing three forward primers that had either an A, T, or G
at the 3ø end (known as AIN1-A, AIN1-T, and AIN1-G, Tables 2
and 3), in conjunction with a conserved reverse primer (3AIN3–62),
samples which contained alleles that differ at nucleotide 75 in the
first intron were separately amplified and sequenced with the dye
terminator method using either an ABI 373 or 377 automated DNA
sequencer (Perkin Elmer/Applied Biosystems, Foster City, CA,
USA). DNA from samples in which both alleles co-amplified with
the same primer set was sequenced using the dye primer method.
Amplified DNA was purified using Microcon-100 columns (Amicon,
Beverly MA, USA) and diluted in 30–100 ml water, according to the
intensity of the amplified band present on an agarose gel stained
with ethidium bromide, prior to the sequencing reaction. Sequence
analysis was performed using the MatchTools software package
(Applied Biosystems).
The HLA-B locus alleles from all samples were characterized by
direct automated DNA sequencing as previously described (9). Alleles were separated into two groups based on their amplification
with two primer sets (Bin1TA-M13F and Bin1CG-M13F), utilizing a
TA/CG dimorphism in intron 1 with a conserved intron 3 reverse
Fig. 1. Map of Africa. The Sahel Corridor is outlined with arrows showing the migration of the Shuwa Arabs and Fulani. The large arrow across
western Africa traces the migration of the Negro tribes. Countries mentioned for allele comparisons are labeled.
292
Tissue Antigens 2000: 56: 291–302
primer (Bin3-M13R). Alleles that were amplified separately were
sequenced with the dye terminator method; alleles co-amplifying
with the same primer set were sequenced concurrently using the
dye primer method.
Ellis et al : Class I alleles in Cameroon, Africa
HLA-A locus SSOP used and their wash temperatures
Probe name
Primer sequences
Sequence (5ø-3ø)
Wash temperature (æC)
AIN1-A(M13F)1
5ø TGT AAA ACG ACG GCC AGT GGG GCG CA[G/A] GAC CCG GGA
1
131R
CGC TCT TGG ACC GCG
50
AIN1-G(M13F)
TGT AAA ACG ACG GCC AGT GGG CGC AGG ACC GGG GG
114R1
GGG TAC CGG CAG GAC
50
AIN1-T(M13F)
TGT AAA ACG ACG GCC AGT GGG [G/A]CG CAG GAC CCG GGT
1
GAC CGG AAC ACA CGG
50
5P2
GGG CGT CGA CGG ACT CAG AAT CTC CCC AGA CGC CGA G
114Q1
TAC CAG CAG GAC GCC
50
ex2R
GCT CTG GTT GTA GTA GCC GC
77S1
GAG AGC CTG CGG ATC
50
ex3F
CCT GGA GAA CGG GAA GGA CAA
76AN1
GAC CGA GCG AAC CTG
50
6N2
GCA GCC TGA GAG TAG CTC CCT
1
1
62RN
62LQ
TGG GAC CTG CAG ACA
50
114EH1
TAT GAA CAG CAC GCC
46
152W1
GCC CGT TGG GCG GAG
50
GAG AGG CCT GAG TAT
50
GCG GCC CGT GTG GCG
56
2
2
56R
1
151R1
73I
1
TCA CAG ATT GAC CGA G
48
142TK1
ACC ACC AAG CAC AAG
50
163R1
GAG GGC CGG TGC GTG
50
156W1
152E
1
GAG CAG TGG AGA GCC
50
GCC CAT GAG GCG GAG
50
149T1
TGG GAG ACG GCC CAT
50
3E207S
GAG GGC ACG TGC GT
48
3E207
GAG GGC GAG TGC GT
48
166DG1
GTG GAC GGG CTC CG
50
M13 sequence is underlined
(8)
Table 2
Alleles amplifying in HLA-A locus primer groups1
AIN-A
AIN-T
AIN-G
A*02
A*29
A*01
A*23
A*31
A*03
A*24
A*32
A*11
A*25
A*33
A*30
A*26
A*36
A*34
A*3204
A*43
161D1
CTG GAT GGC ACG TGC
50
1
GAG CAG CAG AGA GCC
50
150V1
AGG CGG TCC ATG CG
50
142IK1
CAG ATC ACC AAG CGC A
50
62EG1
CGA GGA GAC AGG GAA A
50
62G1
GAC GGG GAG ACA CGG
50
2E162
ACG GAT GTG AAG AAA TAC
50
2E2202
CCC AGG TCC ACT CGG
50
70H
AAG GCC CAC TCA CAG A
50
70Q2
TCT GTG ACT GGG CCT T
50
156L
GGA GCA GTT GAG AGC C
50
74H
ACT CGG TGA GTC TGT G
50
43R
ATC CTC CGG CTC GCG
50
D23
CGG ATC GCG CTC CGC TAC
56
144Q1
AGA TCA CCC AGC GCA
50
A*66
156Q
A*68
A*69
A*80
1
2
Oh et al. (6)
Sequence from non-coding strand.
Table 1
1
Not all alleles tested within family
Table 3
was performed on the amplified template with the same forward
primer (5P2) and a specific reverse primer (ex2R) located in exon 2
that would amplify B*4016, but not B*4701 and B*5802 (the second
HLA-B locus alleles present in individuals CS25 and CS48). A specific forward primer (ex3F) located in exon 3 was used with the reverse primer 6N located in exon 6, to amplify exons 4 and 5 of
B*4016, but not the other HLA-B locus alleles present in individuals
CS25 and CS48. Dye terminator sequencing was performed using
the primers 5P2, ex2R, ex3F, and 6N.
In the case of novel alleles or ambiguous dye primer results,
In addition, exons 1, 4, and 5 of the novel allele HLA-B*4016
amplified DNA was cloned to separate the alleles using the pCRII
were sequenced. To sequence exon 1, a forward primer (5P2) (primer
TA cloning kit (Invitrogen, San Diego, CA, USA). DNA inserts from
sequences are listed in Table 2) located in the 5ø untranslated region
the cloned allele were directly amplified from a lysate containing
was used in conjunction with the reverse primer Bin3-M13R to co-
the bacterial clone heated to 99æC in 50 ml water and sequenced by
amplify both alleles of the HLA-B locus. A second amplification
the dye terminator method. Both strands of DNA preparations from
Tissue Antigens 2000: 56: 291–302
293
Ellis et al : Class I alleles in Cameroon, Africa
multiple polymerase chain reactions (PCR) were sequenced for each
HLA-A alleles and their frequencies (n⫽91)
novel allele.
Number of times observed
The names listed for these sequences have been officially as-
HLA-A allele
Total
DT series
CS series
Total
frequency (%)1
signed by the WHO Nomenclature Committee (10). This follows the
A*0101
2
0
2
1.1
agreed policy that, subject to the conditions stated in the most re-
A*0201/*02092
13
7
6
7.1
cent Nomenclature Report, names will be assigned to new se-
A*0202
15
8
7
8.2
quences as they are identified. Lists of such new names will be
A*0204
1
0
1
0.6
published in the following WHO Nomenclature Report.
A*0205
4
4
0
2.2
Among the novel alleles detected in this study, two of the three
A*0211
1
1
0
0.6
HLA-A locus alleles would not have been detected by common
A*0214
1
0
1
0.6
SSOP methods because the novel sequence substitutions occur at
A*03011
14
8
6
7.7
conserved positions not commonly interrogated by SSOPs. There-
A*030123
fore, there is a possibility that other novel alleles can elude detection
A*2301
by SSOP because their sequence substitutions occur at positions
A*2402/*2409N2
1
0
1
0.5
previously thought to be relatively conserved. An automated se-
A*2601
3
1
2
1.7
quencing approach minimized the number of ‘‘hidden’’ alleles that
A*2612
1
0
1
0.6
would otherwise remain uncharacterized, especially in a population
A*2902
19
6
13
10.4
expected to be diverse.
1
1
0
0.6
34
20
14
18.7
A*3001
10
3
7
5.5
The Ewens-Watterson homozygosity statistic (F) (11, 12) was
A*3002
11
5
6
6.0
used to examine the selective forces influencing allelic diversity of
A*3003
2
0
2
1.1
the Cameroon HLA-A and -B loci. The observed alleleic frequency
A*3004
2
1
1
1.1
distributions were compared to those expected under the neutral
A*3006
1
0
1
0.6
model, in which mutation and random genetic drift are the only
A*3101
2
1
1
1.1
forces altering allele frequencies. The MONTE CARLO program (13,
A*3201
2
2
0
1.1
14) was used to perform the Ewens exact tests and homozygosity
A*3303
5
1
4
2.8
tests on these allele distributions.
A*3402
4
3
1
2.2
2
The exact test of Guo and Thomson (15) was used to evaluate
A*3601
4
0
4
2.2
deviations from expected Hardy-Weinberg genotypic proportions. A
A*6601
11
1
10
6.0
minor deviation from expected Hardy-Weinberg ratios was observed
A*6602
2
1
1
1.1
for the Etoa population at the HLA-B locus. Since these individuals
A*6801
1
1
0
0.6
are from a small population group, it is possible that individuals
A*6802
5
3
2
2.8
are more related than previously thought resulting in this deviation.
A*6901
1
1
0
0.6
A*7401/*74022
9
3
6
5.0
182
82
100
HLA-A-B haplotypes were estimated using standard methods (16,
17).
Total
1
2
Results and discussion
3
HLA-A locus
100
Frequency is defined as the percentage of the number of times the allele was observed in 182
alleles total. Alleles from presumed homozygous samples were counted twice.
Alleles differing outside of exons 2 and 3 were not resolved. A*0209 exists only as a protein
sequence that differs from the protein sequence of A*0201 at one residue in exon 4. The allele
has never been observed a second time.
Novel alleles described in this report are in bold.
Table 4
A total of 30 alleles were present in a sample size of 91 unrelated
individuals. [There was not sufficient DNA to type one of the 92
HLA-A samples.] The most frequently detected alleles were A*2301
(18.7%) and A*2902 (10.4%) (Table 4). HLA-A*0202 (8.2%),
A*2612 and A*3006, which are described in detail below. The allele
A*03011 (7.7%), A*0201 (7.1%), A*6601 (6.0%), A*3002 (6.0%), and
names and Genbank accession numbers for the novel alleles de-
A*3001 (5.5%) were also observed at frequencies greater than 5%.
scribed in this study are listed in Table 5 and their sequences in
Three novel HLA-A alleles were identified, HLA-A*03012,
294
Tissue Antigens 2000: 56: 291–302
Fig. 2.
Ellis et al : Class I alleles in Cameroon, Africa
Table 5
Samples containing novel alleles
Sample
HLA-A
2
HLA-B
Genbank Accession no.1
DT18
A*03012 , A*2301
B*0801, B*3501
AF053128, AF053129
CS3
A*2612, A*2902
B*1503, B*4407
AF065486, AF065487
CS48
A*3006, A*0202
B*4016, B*5802
AF028713, AF028714 (A*3006)
AF027296, AF027297 (B*4016)
CS25
A*0202, A*3004
B*4016, B*4701
AF017022, AF017023
DT16
A*0201, A*2301
B*1403, B*4501
U91330, U91331
DT3
A*2301
B*1403, B*5801
AF015271, AF015272
DT23
A*2301, *3004
B*1403, B*3501
–
DT32
A*0201, A*0202
B*4703, B*44031
AF016842, AF016843
1
2
Novel allele only, corresponds to two exons.
New alleles described in this report are bolded.
Fig. 2. Exon 2 and 3 nucleotide sequences of the three novel
HLA-A alleles are compared to the allele with closest homology.
Numbering begins at the first nucleotide of exon 1 and disregards introns
(18).
Tissue Antigens 2000: 56: 291–302
295
Ellis et al : Class I alleles in Cameroon, Africa
B*5802 (both 10.9%). HLA-B*5301 is the most frequently detected
HLA-B alleles and their frequencies (n⫽92)
allele in other West African populations and has been hypothesized
Number of times observed
HLA-B allele
Total
DT series
CS series
Total frequency (%)1
to be associated with resistance to severe malaria (19). Malaria is
B*07021
11
6
5
6.0
endemic to the region of Cameroon where these samples originated.
B*0801
10
7
3
5.4
HLA-B*3501 (7.1%), *44031 (6.0%), *07021 (6.0%), *4901 (5.4%),
B*1302
5
1
4
2.7
*5801 (5.4%) and *0801 (5.4%) were also frequent in this popula-
B*1401
2
1
1
1.1
tion. Three novel HLA-B alleles were characterized, HLA-B*1403,
B*1402
1
0
1
0.5
*4016, and *4703, which are described in detail below and in Table
B*14032
3
3
0
1.6
5 and Fig. 3.
B*1501
2
0
2
1.1
B*1503
9
5
4
4.9
B*1510
2
1
1
1.1
B*1516
2
2
0
1.1
B*1517
2
1
1
1.1
The Ewens-Watterson homozygosity statistic (F) has been used to
B*1801
5
1
4
2.7
infer the operation of evolutionary forces on allele frequency distri-
B*2703
1
1
0
0.5
butions (20–23). In general, homozygosity values lower than the
B*3501
13
4
9
7.1
neutral value predicted by the Ewens-Watterson model reflect the
B*3701
1
0
1
0.5
operation of balancing selection, while values higher than the neu-
B*3910
1
0
1
0.5
tral value reflect the influence of directional selection. By comparing
B*4016
2
0
2
1.1
observed homozygosity values with those expected under this
B*4101
1
0
1
0.5
model, Klitz et al. (20) have shown that balancing selection (hetero-
B*4201
9
5
4
4.9
zygote advantage) is in operation at the HLA-DRB, -DQA1 and -
B*44031
11
7
4
6.0
DQB1 loci. While this phenomenon has been observed in a large
B*44032
1
0
1
0.5
number of populations at these class II loci, it has not been demon-
B*4407
4
0
4
2.2
strated as clearly for the class I loci (22).
B*4501
6
4
2
3.3
In particular, the homozygosity value for the HLA-A locus in
B*4701
2
1
1
1.1
the Yaoundé population is significantly lower than expected for a
B*4703
1
1
0
0.5
population of the same size with the same number of alleles (Table
B*4901
10
5
5
5.4
7), and is consistent with the operation of balancing selection. In
B*5001
1
0
1
0.5
addition, an overall trend toward lower observed homozygosity
B*51011
3
2
1
1.6
values is evident for the Cameroon HLA-A and -B loci, consistent
B*5301
20
6
14
10.9
with the observations of Klitz et al. (20) for the class II loci, sug-
B*5703
5
3
2
2.7
gesting that balancing selection is operating on these populations,
B*5801
10
5
5
5.4
perhaps in concert with other evolutionary forces.
B*5802
20
6
14
10.9
8
4
4
4.4
184
82
102
B*8101
Total
1
2
Allele diversity and selective forces
100
Frequency is defined as the percentage of the number of times allele present in 184 alleles
total. Alleles for presumed homozygous samples are counted twice.
Novel alleles described in this report are in bold.
Table 6
Novel allele – HLA-A*03012
DNA from samples typed by SSOP as A*0301 or A*0303N was
sequenced to differentiate between these alleles. HLA-A*0303N contains a deletion in exon 3. Although the null allele was not detected,
a single nucleotide variant of A*0301 was identified. The codon
CTG at 156 was present in A*03012, resulting in a silent substi-
HLA-B locus
tution and no predicted amino acid change between A*03011 and
Thirty-three alleles were identified in 92 unrelated individuals
A*03012. Only A*8001 and A*03012 have CTG at codon 156, al-
(Table 6). The most frequently observed alleles were B*5301 and
though many HLA-B locus alleles contain this sequence.
296
Tissue Antigens 2000: 56: 291–302
Ellis et al : Class I alleles in Cameroon, Africa
Fig. 3. Exon 2 and 3 nucleotide sequences of the novel alleles
HLA-B*1403 and B*4703 are compared to the allele with closest
homology. Exons 1–6 of HLA-B*4016 are compared to both HLA-B*40012
and B*0706. Numbering begins at the first nucleotide of exon 1 (18).
Tissue Antigens 2000: 56: 291–302
297
Ellis et al : Class I alleles in Cameroon, Africa
Table 7
Homozygosity in Cameroon populations
Locus
Sample series
A
DT (Yaoundé)
A
CS (Etoa)
B
DT (Yaoundé)
B
CS (Etoa)
1
2n
k
Observed F
Expected F
p-value
82
24
0.0577
0.0901
⬍0.005
102
29
0.0668
0.0768
n.s.1
82
22
0.1047
0.1013
n.s.
100
24
0.0764
0.0993
n.s.
n.s., not significant
later identified in 8 unrelated individuals from another study, 7 of
Novel allele – HLA-A*2612
whom were African American and one of Middle Eastern descent
A single nucleotide variant of A*2601 was characterized from one
(data not shown). Based on this, HLA-B*1403 might be frequent in
individual in the CS series with a variant SSOP pattern. Probe 152E
African populations.
(covering nucleotides 520–534 in exon 3) did not hybridize, resulting
in an assigned typing of A*26v, A*2901/*2902. SBT demonstrated
HLA-A*2612 contains a GTG codon at 156, rather than GAG pres-
Novel allele – HLA-B*4703
ent in HLA-A*2601. A single amino acid change is predicted al-
HLA-B*4703 was identified from one individual and, later, 3 unre-
tering glutamic acid in A*2601 to valine in A*2612. Other HLA-A
lated African American individuals (data not shown). A two nucleo-
alleles, such as A*0201, contain this A*2612 sequence motif.
tide variant of B*4702, two amino acid substitutions at residues 82
(arginine to lysine) and 83 (glycine to arginine) are predicted in the
polypeptide encoded by B*4703. Codons 77–83 encode the Bw4 and
Novel allele – HLA-A*3006
Bw6 epitopes. The predicted polypeptide sequence is a novel combi-
A single nucleotide variant of HLA-A*3004, A*3006, was identified
nation containing residues 77 through 81 of the Bw6 epitope
from one individual. The GCG at codon 31 of the mature protein is
(SLRNL) joined to the last two residues of the Bw4 epitope (LR)
unique; all HLA-A alleles have the codon ACG, with the exception
(Fig. 4). No other alleles contain this hybrid sequence. The Class Ib
of A*8001 which has TCG at this position (24). This codon has not
gene, HLA-F, and the chimpanzee homologue of HLA-F contain the
been observed in any other HLA class I alleles, pseudogenes, primate alleles, or primate pseudogene sequences examined (25–29).
Using SSOP, this new allele would not have been identified as a
variant due to an absence of probes in the generally conserved region around codon 31. Due to the presence of a novel HLA-B allele
(B*4016) in this sample, the alleles of the HLA-A locus were also
sequenced rather than typed by SSOP. DNA from the other two
individuals typed as HLA-A*3004 by SSOP was also sequenced
confirming that they were indeed A*3004, not A*3006.
Novel allele – HLA-B*1403
A single nucleotide variant of HLA-B*1402 was identified from
DNA from individuals in the DT Series. HLA-B*1403 contains the
sequence CGG (rather than CTG in *1402) at codon 156, resulting
in a predicted amino acid change from leucine (B*1402) to arginine
(B*1403). This sequence is present at codon 156 in other reported
alleles, such as B*0705. A total of three individuals possessed the
novel B*1403 allele; all B*1403 positive individuals also carried
A*2301, suggesting an association with B*1403. HLA-B*1403 was
298
Tissue Antigens 2000: 56: 291–302
Fig. 4. Protein sequences observed at the Bw4/Bw6 epitope (codons 77–83) (26, 30).
Ellis et al : Class I alleles in Cameroon, Africa
same predicted sequence as HLA-B*4703 spanning codons 78–83,
but differ at codon 77 (alanine rather than the serine present in
B*4703) (26, 31). The chimpanzee allele, Patr-B*04, also contains
the same Bw4/Bw6 hybrid, with the exception of a glycine (G) at
codon 77. Studies of the Bw4 and Bw6 epitopes reveal that residues
77 and 80–83 are necessary components of these epitopes with residues 82 and 83 being the most crucial. We have predicted that this
novel HLA-B locus molecule bears a Bw4 serologic epitope since
R82 and G83 are essential for the Bw6 epitope and studies with a
B7/B27 hybrid generated by mutagenesis containing the SLRNLLR
sequence seen in HLA-B*4703 was shown to be Bw4 reactive (8,
32). Studies by Darke et al. have shown that HLA-B*4703 is reactive
with most or all Bw4 and some Bw6 antisera (33).
Novel allele – HLA-B*4016
HLA-B*4016 differs from B*40012 by five nucleotides, with each
substitution yielding a predicted amino acid change. The exons 1,
2, 4, and 5 of this allele are identical to B*40012, while exons 3, 4,
and 5 are identical to B*0705 and B*0706. Exon 4 and 5 sequences
of B*40012, *4016, *0705, and *0706 are identical. B*4016 possibly
arose following a reciprocal recombination event or an intraallelic
conversion event involving B*0705 (or *0706) and B*40012. B*4016
was present in two unrelated Cameroonian individuals, both from
Estimated haplotypes1 in Cameroon
Haplotype (A-B)
Frequency
Haplotype (A-B)
Frequency
0201–07021
0.0110
2902–4407
0.0110
0201–1801
0.0165
2902–4901
0.0220
0210–5301
0.0110
2902–5301
0.0220
0202–4016
0.0110
2902–5802
0.0110
0202–44031
0.0165
2902–8101
0.0110
0301–0801
0.0110
3001–4201
0.0165
0301–4201
0.0110
3001–5301
0.0110
0301–5301
0.0110
3001–5802
0.0110
0301–5801
0.0110
3002–1302
0.0165
0301–5802
0.0110
3002–5301
0.0165
2301–07021
0.0165
3002–5801
0.0110
2301–1403
0.0110
3303–5802
0.0110
2301–3501
0.0330
3402–51011
0.0110
2301–44031
0.0110
3601–5301
0.0165
2301–4501
0.0110
6601–3501
0.0110
2301–5301
0.0110
6601–5802
0.0110
2301–5801
0.0110
6802–0801
0.0110
2301–5802
0.0110
7401/2–1503
0.0110
2301–8101
0.0165
7401/2–5802
0.0110
2601–07021
0.0110
1
Only haplotypes appearing more than once are listed
Table 8
the village of Etoa and later identified in six other unrelated African
American individuals (data not shown). The sequence of this allele
was erroneously assigned both names HLA-B*4016 and HLAB*4017 by the WHO Nomenclature Committee and the name HLA-
Comparison of populations
B*4017 was dropped.
Few class I allele level population studies are available, making it difHLA-A and HLA-B associations
ficult to compare class I frequencies among populations. For this
study of the HLA-A locus, alleles have been grouped into predicted
A large number of HLA-A-B haplotypes were observed in both Ca-
serologic groups to compare to existing serologic level population
meroon population samples. Most haplotypes were observed once
studies. Fig. 5 compares the frequencies of the HLA-A antigens in Ca-
in each population sample, as indicated by the high value of diver-
meroon to the 1991 International Histocompatibility Workshop sero-
sity (h) for this population (h⫽0.98). Although not significantly as-
logic results for ‘‘West Africans’’ (34). A74 and A66 were not tested
sociated at the antigen level, the A*2301-B*3501 haplotype was the
for in the Workshop but these types would have been identified as
most common Cameroon haplotype, ocurring at a frequency of ap-
A19 and A10. A66 is also often typed A26 (35). A74 and A66 are listed
proximately 3–4% overall. Predicted haplotypes occurring more
in Fig. 5 for purposes of showing the frequency in Cameroon.
than once in the population are listed in Table 8.
Absent in both the Cameroon and Workshop results are HLA-
It seems likely that the pool of HLA-A-B haplotypes in the Ca-
A11, A25, and A43. HLA-A36 was not detected in the Workshop,
meroon population is very large, due both to the extensive allelic
but is present in Cameroon. The frequency of HLA-A29 is dramati-
diversity at these loci, and the observation that these allele levels
cally higher in Cameroon (10.4% vs. 2.3%), the frequency of HLA-
are maintained at relatively even levels by balancing selection. Sub-
A23 is nearly double (18.7% vs. 9.7%) and the frequency of HLA-
sequent studies of this and other west African populations should
A3 is increased (8.2% vs. 1.5%). Conversely, there is a much lower
endeavor to obtain very large sample sizes so as to more accurately
frequency in Cameroon of HLA-A1 (1.1% vs. 5.3%), HLA-A24 (0.6%
assess the extent and distribution of HLA-A-B haplotype diversity.
vs. 3.8%), HLA-A28 (3.9% vs. 12.6%), and HLA-A33 (2.7% vs.
Tissue Antigens 2000: 56: 291–302
299
Ellis et al : Class I alleles in Cameroon, Africa
Fig. 5. HLA-A antigen frequency comparison between the International Histocompatibility Workshop West Africans (34) and
Cameroon. HLA-A74 and -A66 were not tested for by workshop serology.
BL is HLA-A blank. Frequencies in Cameroon are represented by the
light bars (left), while the West African frequencies are the solid bars
(right).
15.2%). The most frequent serologic type, HLA-A2, has a similar
likely positively selected in an evolutionary timeframe. Whether
frequency in both populations (19.2% vs. 20.6%). Within the HLA-
other alleles frequent in this population have been positively se-
A2 family, 6 of the 30 known A2 alleles were detected in Cameroon,
lected for their ability to efficiently present peptides from patho-
including A*0201, A*0202 , A*0204, A*0205, A*0211, and A*0214.
genic organisms is unknown.
The diversity at the HLA-B locus makes these comparisons dif-
The inference of balancing selection in the Etoa population and
ficult because so many recently described HLA-B alleles have been
the generally low values of F for the Cameroon HLA-A and -B loci
defined solely by sequencing and not by serology. Of interest, and
suggests that the class I loci in this population have been subject
probably the most relevant, is the variation in HLA-B53 frequency
to evolutionary forces similar to those observed at the class II loci.
in western Africa. In Cameroon, HLA-B*5301 was present at a fre-
Similarly low homozygosity values have been observed at the HLA-
quency of 10.9%. In other African populations, HLA-B53 was sero-
A and B loci in Pacific/Asian populations, and a significantly low
logically detected in 22.2% of Nigerians, and 13.9% of the Serere
value has been observed in Chinese populations (22), suggesting
and Mandinka in Senegal, and 14–17% of two tribes in The Gambia
that class I and class II allele frequency distributions in populations
(Wolof and Mandinka), 8.8% of Zaireans, 8.3% of Zimbabweans,
around the globe have been shaped by balancing selection.
1.0% of South African Xhosa (36, 37). The variation in HLA-B53
frequency between different African populations could reflect the
Allelic diversity
diverse genetic backgrounds of these groups. The many historical
migrations of different tribes and peoples through Cameroon due to
A diverse group of alleles was detected in Cameroon. In a population
war, famine, expansion, colonization, and emigration, is reflected in
of slightly less than a hundred individuals, 30 HLA-A locus alleles
the allelic diversity at the HLA-A and HLA-B loci. This diversity
and 33 HLA-B locus alleles were detected. Diversity values (h) for the
may have also arisen under selective pressure as Cameroon is en-
HLA-A loci of the Etoa and Yaoundé population samples are 0.95 and
demic for P. falciparum malaria, among other parasites. The higher
0.97, respectively, while the values for the HLA-B loci are 0.94 and
frequency of HLA-B*5301, which has been hypothesized to provide
0.92. In comparison, the HLA-A locus diversity in Pacific/Asian popu-
resistance to severe malaria, is an example of an allele that was
lations is uniformly lower (h⫽0.37–0.88)(22). Comparing Camerooni-
300
Tissue Antigens 2000: 56: 291–302
Ellis et al : Class I alleles in Cameroon, Africa
ans, descending from a large founder population, to South American
to 50,000 per year during the peak of the slave trade in the 18th
Indians, who are believed to have originated from a much smaller
century. Most of the exported slaves arrived in the Americas and,
founder population, there was a drastic difference in the number of
thus, alleles present in West African populations are present in
alleles detected in the populations. As few as 3 to 6 HLA-A and 6 to 20
African Americans. All novel HLA-B alleles described in this study
different HLA-B alleles have been detected in South American Indian
of Cameroon were later identified in several unrelated African
populations with similar to larger sample sizes than our current study
Americans (and, in the case of HLA-B*1403, one individual of
(38). This observation supports the impact of founder population size
Middle Eastern origin) in routine typing of patients and volunteer
on the allelic diversity present in different populations.
donors for an American bone marrow registry. Allele level population studies of the HLA class I loci in Africans provide helpful
frequency data to be used in the development of stem cell registries
Alleles in populations of African origin
and the development of subunit or peptide-based vaccines, in adDuring the sixteenth century, between one and two thousand slaves
dition to aiding the understanding of human evolution and the im-
of West African origin were exported annually. This number rose
pact of disease on HLA polymorphism.
References
1. Parham P, Ohta T. Population biology of
antigen presentation by MHC class I
molecules. Science 1996: 272: 67–74.
2. Nei M, Livshits G. Genetic relationships of
Euroasians, Asians, and Africans and the
origin of modern Homo sapiens. Hum
Heredity 1989: 39: 276–81.
3. Stine OC, Dover GJ, Zhu D, Smith KD. The
evolution of two West African
populations. J Mol Evol 1992: 34: 336–44.
4. McEvedy, C. The Penguin Atlas of African
History. 2nd edition. New York: Penguin
Books, 1995.
5. Nelson, HD. Area Handbook for the United
Republic of Cameroon. 1st edition.
Washington, DC: U.S. Government Printing
Office, 1974.
6. Oh S, Fleischhauer K, Yang SY. Isoelectric
focusing subtypes of HLA-A can be defined
by oligonucleotide typing. Tissue Antigens
1993: 41: 135–42.
7. Cereb N, Maye P, Lee S, Kong Y, Yang SY.
Locus-specific amplification of HLA class
I genes from genomic DNA: locus-specific
sequences in the first and third introns of
HLA-A, -B, and -C alleles. Tissue Antigens
1995: 45: 1–11.
8. Lutz CT, Smith KD, Greazel NS et al. Bw4reactive and Bw6-reactive antibodies
recognize multiple distinct HLA structures
that partially overlap in the alpha-1 helix.
J Immunol 1994: 153: 4099–110.
9. Steiner N, Ng J, Bush J, Hartzman RJ,
Johnston-Dow L, Hurley CK. HLA-B alleles
associated with the B15 serologically
defined antigens. Hum Immunol 1997: 56:
84–93.
10. Bodmer JG, Marsh SGE, Albert ED et al.
Nomenclature for factors of the HLA
system. Tissue Antigens 1999: 53: 407–46.
11. Ewens WJ. The sampling theory of
selectively neutral alleles. Theor Popul Biol
1972: 3: 87–112.
12. Watterson GA. The homozygosity test of
neutrality. Genetics 1978: 88: 405–17.
13. Slatkin M. An exact test for neutrality
based on the Ewens sampling distribution.
Genet Res 1994: 64: 71–4.
14. Slatkin M. A correction to the exact test
based on the Ewens sampling distribution.
Genet Res 1996: 68: 259–60.
15. Guo SW, Thompson EA. Performing the
exact test of Hardy-Weinberg proportion for
multiple alleles. Biometrics 1992: 48: 361–
72.
16. Baur MP, Danilovs JA. Population analysis
of HLA-A, B, C, DR and other genetic
markers. In: Terasaki PI, ed.
Histocompatibility Testing 1980. Los
Angeles: UCLA Press, 1980.
17. Long JC, Williams RC, Urbanek M. An EM algorithm and testing strategy for
multiple locus haplotypes. Am J Hum
Genet 1995: 56: 799–810.
18. Mason PM, Parham P. HLA class I and II
sequence alignments.
www.anthonynolan.com/HIG/data.html
1999.
19. Hill AVS, Allsopp CEM, Kwiatkowski D et
al. Common West African HLA antigens
are associated with protection from severe
malaria. Nature 1991: 352: 595–600.
20. Klitz W, Thomson G, Baur MP. Contrasting
evolutionary histories among tightly linked
HLA loci. Am J Hum Genet 1986: 39: 340–
9.
21. Titus-Trachtenberg EA, Rickards O, De
Stefano GF, Erlich HA. Analysis of HLA
class II haplotypes in the Cayapa Indians
of Ecuador: a novel DRB1 allele reveals
evidence for convergent evolution and
balancing selection at position 86. Am J
Hum Genet 1994: 55: 160–7.
22. Bugawan TL, Mack SJ, Stoneking M et al.
HLA class I allele distributions at six
Pacific/Asian populations: evidence of
selection at the HLA-A locus. Tissue
Antigens 1999: 53: 311–9.
23. Mack SJ, Bugawan TL, Moonsamy PV et
al. Evolution of Pacific/Asian populations
inferred from HLA class II allele frequency
distributions. Tissue Antigens 2000: 55:
383–400.
24. Wagner AG, Hughes AL, Iandoli ML et al.
HLA-A*8001 is a member of a newly
discovered ancient family of HLA-A alleles.
Tissue Antigens 1993: 42: 522–9.
Tissue Antigens 2000: 56: 291–302
301
Ellis et al : Class I alleles in Cameroon, Africa
25. Cadavid LF, Hughes AL, Watkins DI. MHC
class I-processed pseudogenes in New World
primates provide evidence for rapid
turnover of MHC class I genes. J Immunol
1996: 157: 2403–9.
26. Lawlor DA, Warren E, Ward FE, Parham P.
Comparison of class I MCH alleles in
humans and apes. Immunol Rev 1990: 113:
147–85.
27. Watkins DI, Zemmour J, Parham P. Nonhuman primate MHC class I sequences.
Immunogenetics 1993: 37: 330.
28. Geraghty DE, Koller BH, Pei JI, Hansen JA.
Examination of four HLA class I
pseudogenes. J Immunol 1992: 149: 1947–
56.
29. Suarez B, Morales P, Castro MJ et al. MHCE polymorphism in Pongidae primates: the
same allele is found in two different
species. Tissue Antigens 1997: 50: 695–8.
30. Parham P, Lawlor DA. Evolution of class I
major histocompatibility complex genes
and molecules in humans and apes. Hum
Immunol 1991: 30: 119–28.
302
31. Allsopp CEM, Hill AVS, Kwiatkowski D et
al. Sequence analysis of HLA-Bw53, a
common West African allele, suggests an
origin by gene conversion of HLA-B35. Hum
Immunol 1991: 30: 105–9.
32. Toubert A, Raffoux C, Boretto J et al.
Epitope mapping of HLA-B27 and HLA-B7
antigens by using intradomain
recombinants. J Immunol 1988: 141:
2503–9.
33. Darke C, Guttridge MG, Street J, Thompson
J, Thomas M. HLA-B*4703: sequence
confirmation, serology and distribution.
Tissue Antigens 1999: 53: 586–90.
34. Imanishi T, Akaza T, Kimura A, Tokunaga
K, Gojobori T. Allele and haplotype
frequencies for HLA and complement loci
in various ethnic groups. In: Tsuji K,
Aizawa M, Sasazuki T, eds. HLA 1991
Proceedings of the Eleventh International
Histocompatibility Workshop and
Conference. Oxford: Oxford University
Press, 1992: 1065–220.
Tissue Antigens 2000: 56: 291–302
35. Hurley CK, Schreuder GMT, Marsh SGE,
Lau M, Middleton D, Noreen H. The search
for HLA-matched donors: a summary of
the HLA-A*, -B*, -DRB1*1/3/4/5* alleles
and their association with serologically
defined HLA-A, -B, and -DR antigens. Tissue
Antigens 1997: 50: 401–18.
36. Halle L, Mbayo K, Lurhuma Z et al. HLAA, B, C, DR and DQ polymorphism in
Zarieans. Tissue Antigens 1994: 44: 196–9.
37. Dieye A, Diaw ML, Rogier C, Trape JF,
Sarthou JL. HLA-A, -B, -C, -DR, -DQ typing
in a population group of Senegal:
distribution of HLA antigens and HLADRB1*13 and DRB1*11 subtyping by PCR
using sequence-specific primers (PCR-SSP).
Tissue Antigens 1996: 47: 194–9.
38. Cadavid LF, Watkins DI. Heirs of the jaguar
and the anaconda: HLA, conquest and
disease in the indigenous populations of
the Americas. Tissue Antigens 1997: 50:
702–11.