Predisposition to idiopathic thrombocytopenic purpura maps

Predisposition to idiopathic thrombocytopenic purpura maps

Human Immunology 70 (2009) 179-183
Contents lists available at ScienceDirect
Predisposition to idiopathic thrombocytopenic purpura maps close to the major
histocompatibility complex class I chain-related gene A
Maria Helena Thomaz Maia a,*, Raquel de Lima Peixoto a, Clayton Pereira Silva de Lima a,
Milena MagalhÄes a, Leonardo Sena b, PerpÊtua do Socorro Silva Costa a, FabÎola Brasil Barbosa a,
Layanna Freitas de Oliveira a, Matilde Romero c, Christina Nogueira de Araujo Silva c,
Eduardo JosÊ Melo dos Santos a
a
b
c
Laboratòrio de Genètica Humana e Mèdica, Instituto de Ciéncias Biològicas, Universidade Federal do Parà, Belèm, PA, Brasil
Departamento de Patologia, Universidade do Estado do Parà, Belèm, PA, Brasil
Laboratòrio de Imunogenètica, Instituto Nacional do Cáncer, Rio de Janeiro, RJ, Brasil
A R T I C L E
I N F O
Article history:
Received 26 May 2008
Accepted 16 January 2009
Available online 23 January 2009
Keywords:
Idiopathic thrombocytopenic purpura
MHC
Microsatellites
MICA
A B S T R A C T
Idiopathic thrombocytopenic purpura (ITP) is an autoimmune condition with poorly known etiology, characterized by platelet destruction. Genetic association studies of this disease are scarce, discrepant, and
restricted to major histocompatibility complex (MHC) polymorphisms. Hence, a case– control study was
conducted with an aim to map the MHC to IPT susceptibility using HLA-B and nine microsatellite loci
encompassing MHC class I, II, and III regions. We compared the allelic frequencies in samples of unrelated
healthy controls and ITP patients. After correction for multiple tests, only allele MICA*183, also known as
A5.1, demonstrated an association, resulting in the identification of a major predisposing region close to
STR-MICA. This result may highlight the putative functional role of MICA in the immune response to ITP.
䉷 2009 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights
reserved.
1. Introduction
Autoimmune thrombocytopenia arises from the early destruction of platelets by antiplatelet autoantibodies in the activated
endothelial reticulum system [1,2]. The resulting disease is characterized by a severe decrease in platelet numbers in the peripheral
blood, followed by symptoms such as petechiae, purpura, conjunctional hemorrhage, or other types of mucocutaneous bleeding [3].
Autoantibodies found in autoimmune thrombocytopenia may have
distinct origins, differentiating the condition into four broad
groups: idiopathic thrombocytopenic purpura (ITP), secondary immune thrombocytopenia (e.g., secondary to systemic lupus erythematosus), drug-induced immune thrombocytopenia, and viral
infection-related thrombocytopenia (e.g., HIV) [1]. Among those
groups, ITP has a significant incidence of around 6:100,000 inhabitants per year [4], although the fundamental disturbances that
lead to its autoimmune response are unknown. Genetic factors can
influence the development of autoimmune diseases, especially
immune-response genes, including the human leukocyte antigen
(HLA) genes. Indeed, some association studies have indicated genetic polymorphisms on HLA genes as predisposing factors to ITP
[5–10]. However, the associations reported were generally weak
and varied among the studies. Possible reasons for the lack of
* Corresponding author.
E-mail address: [email protected] (M.H.T. Maia).
reproducibility in previous studies include different diagnostic criteria for ITP, ethnic variability in the HLA allele distribution, and the
use of serological typing methods, which are unable to identify HLA
alleles at the amino acid level. In addition, HLA analysis may be
further complicated by linkage disequilibrium with other genes of
the major histocompatibility complex (MHC).
The first stage of the present study aimed to map the MHC using
microsatellite markers encompassing the most relevant class I, II,
and III recombination blocks [11] to identify the primary ITP predisposing/protective regions. Because the region around major histocompatibility complex class I chain-related gene A (MICA), represented by the STR-MICA, exhibited the strongest association
signal in our study and the HLA-B locus was reported to be associated with ITP in some studies [8,10], this locus was genotyped and
analyzed to certify that the possible association of ITP with MICA
polymorphisms was not secondary based on linkage disequilibrium with this locus.
2. Subjects and methods
Of about 400 patients registered at the Foundation Center of
Hemotherapy and Hematology of Par (a reference state hematology center in northern Brazil) as autoimmune thrombocytopenia
carriers, only 51 (11 men and 40 women, mean age of 35 years old
at examination) could be diagnosed as adult ITP carriers using the
absence of background diseases and the presence of GPIIb/IIIa
and/or GPIb/IX serum autoantibodies as criteria (PakAuto Kit, GTI,
0198-8859/09/$32.00 - see front matter 䉷 2009 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
doi:10.1016/j.humimm.2009.01.011
180
M.H.T. Maia et al. / Human Immunology 70 (2009) 179-183
Brookfield, WI, USA). The control sample was composed of 145
unrelated individuals (65 men and 80 women, mean age of 31 years
old) representative of the same population from which the ITP
carriers were obtained. Men and women in control sample underwent allelic frequency comparison to determine whether their
lower female/male ratio could bias the results. DNA samples were
isolated according to Sambrook et al. [12].
The nine microsatellites used for genetic mapping (Table 1)
were located in the following MHC recombination blocks: ␧
(D6S2749 locus) and ␦ (G51152 and D6S2883) blocks in the class II
region; in-between ␥ and ␦ blocks (D6S273 and BAT2GT) and ␤ block
(TNFd) in the class III region; and ␤ (STR-MICA and D6S2811) and ␣
(D6S510) blocks in the class I region (Figure 1) [11,13,14].
Polymerase chain reaction (PCR) conditions for each loci were
94⬚C for 10 minutes, followed by 30 cycles at 94⬚C for 30 seconds,
primer-specific annealing temperature for 30 seconds, 72⬚C for 1
minute, and final extension at 72⬚C for 1 hour. BAT2GT and D6S2883
were multiplexed at an annealing temperature of 60⬚C, whereas
D6S2749, G51152, D6S273, and D6S2811 were multiplexed at an
annealing temperature of 57⬚C. Annealing temperatures for TNFd,
STR-MICA, and D6S510, loci that could not be multiplexed, were 64,
57, and 65⬚C, respectively. The final volume of each reaction was 15
␮L, composed of reaction buffer (20 mM Tris–HCl, pH 8.4, 50 mM
KCl), 3.3 mM MgCl2, 0.75 ␮L DMSO, 0.08 mM dNTP, 1 U Platinum
Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA), and 50 ng of
genomic DNA. Primer concentrations were 0.25 mM for D6S510,
TNFd, G51152, and MICA, 0.35 mM for D6S2883, 0.15 mM for
BAT2GT, 0.1 mM D6S2749, 0.2 mM for D6S273, and 0.3 mM for
D6S2811. Forward primers (Table 1) were fluorescently tagged at
the 5= ends, allowing genotyping using an ABI 377 automated
sequence analyzer (Applied Biosystems, Foster City, CA, USA).
HLA-B genotyping was performed using the INNO-LiPA HLA-B
Update Plus probe assay (Innogenetics, Gent, Belgium) according to
the manufacturer’s instructions.
3. Results
Hardy–Weinberg genotypic deviations were calculated using
GENEPOP software [16]. Genetic diversities (H) for each locus were
estimated according to Nei [17] and compared using Wilcoxon’s
test. The ␹2 test was applied for each allele and the p value was
accessed by Monte Carlo simulations, using the CLUMP software
[18]. Odds ratios (OR) and 95% confidence intervals (CI) were estimated for alleles that significantly differed in their frequencies
between patients and controls. Multiple test correction was applied by multiplying allelic p values by the number of alleles at each
Table 1
Characteristics of the nine selected MHC microsatellites
Loci
Primer 1
Primer 2
Core motif
References
D6S2749
GAGGTAATGTCACAGGATGGG
TGCTTATAGGGAGACTACCG
GGTAAAATTCCTGACTGGCC
GACAGCTCTTCTTAACCTGC
ACATTTGTATGCTTCAGATG
TGGAATCTCATCAAGGTCAG
ACCAAACTTCAAATTTTCGGC
GGAGAAGTTGAGTATTTCTG
CTCCAGCCTGGATAACAG
ACAAGGGCTTTAGGAGGTCT
CATAGTGGGACTCTGTCTCCAAAG
AGATCCTTCCCTGTGAGTTCTGCT
CCTTTTTTTCAGGGAAAGTGC
CCTTACCATCTCCAGAAACTGC
TGCCATTTGGCCCTAAATGC
TGGGCAATGAGTCCTATGAC
AATGGGCTACTACTTCACACC
CAACACACTGATTTCCATAGC
GT
13
GT
13
G51152
D6S2883
D6S273
BAT2GT
TNFd
STR-MICA
D6S2811
D6S510
AC
13
GT
13
GT
13;14
AG
13
GCT
13;15
GT
13
AC
13
Fig. 1. Localization map of the microsatellites in the MHC. The nine microsatellites
studied (gray boxes) are shown in their relative position to relevant MHC genes
(black boxes).
locus according to Bonferroni’s correction linkage disequilibrium
and haplotypes were inferred using Pypop software [19].
All loci were in Hardy–Weinberg equilibrium, and male and
female control subsamples did not differ in terms of their allelic
frequencies (␹2 ⫽ 26.618; p ⫽ 0.1464). Microsatellite genetic diversity (H) and number of alleles (k) did not differ statistically between
patients and controls (Wilcoxon’s test: Z ⫽ 1.481, p ⫽ 0.139, for H
comparison; Z ⫽ 1.782, p ⫽ 0.075, for k; Table 2); in addition, the
Table 2
Genetic variability estimates for each locus in controls and patients samples
Loci
D6S2749
G51152
D6S2883
D6S273
BAT2GT
TNFd
STR-MICA
D6S2811
D6S510
HLA-B
Average
Control (2n ⫽ 288)
Patients (2n ⫽ 102)
H
k
H
k
0.744
0.707
0.850
0.779
0.789
0.666
0.772
0.930
0.846
0.931
0.800
08
10
10
08
13
08
05
21
13
27
12.3
0.760
0.759
0.815
0.780
0.822
0.726
0.774
0.940
0.837
0.944
0.816
06
09
10
08
14
06
05
21
09
26
11.4
H ⫽ expected heterozygosity; k ⫽ number of alleles; 2n ⫽ number of chromosomes
sampled.
M.H.T. Maia et al. / Human Immunology 70 (2009) 179-183
181
Fig. 2. Allelic frequencies of all loci studied among patients and controls. (A) D6S2749; (B) G51152; (C) D6S2883; (D) D6S273; (E) BAT2GT; (F) TNFd; (G) STR-MICA; (H)
D6S510; (I) D6S2811; and (J) HLA-B. Black bars ⫽ controls; empty bars ⫽ patients.*p ⬍ 0.05; **pc ⬍ 0.05. Y axis ⫽ allelic frequencies; X axis ⫽ allelic denomination.
values of H and k were similar to those described for other populations, especially those of European origin, and all microsatellite
allele sizes observed were within the allele size range described in
previous studies [13,15,20,21].
The highest difference among patients and controls with regard
to allelic frequencies was observed in the STR-MICA for allele 183
(15%), as shown in Figure 2. The OR and 95% CI for all alleles that
significantly differed in their frequencies between patients and
controls (Table 3) and the most significant OR were obtained for
STR-MICA (183 allele, also known as A5.1 [15]). The most significant
p values among all alleles for each locus were obtained for STR-
MICA (p ⫽ 0.0032), TNFd (p ⫽ 0.0069), BAT2GT (p ⫽ 0.0244), and
D6S2883 (p ⫽ 0.034) loci (Table 3); however, after multiple test
corrections, only STR-MICA, located in the ␤ recombination block,
remained significant (pc ⫽ 0.016). Hence, we believe that the trend
represents an important clue to mapping ITP predisposing genes.
Because HLA-B is a key locus near MICA, which is associated with
several autoimmune conditions, including ITP itself, it constitutes a
predisposing candidate gene. Hence, genotyping of this locus was
performed and its allele frequencies are given in Figure 2. A total of
28 different HLA-B alleles were observed in both samples, 26 in
patients and 27 HLA-B alleles in controls (Table 2). HLA-B*08 allelic
182
M.H.T. Maia et al. / Human Immunology 70 (2009) 179-183
Table 3
p values and odds ratio estimates for all loci
Locus
Allele
OR
p values
pc values
D6S2749
G51152
D6S2883
D6S273
BAT2GT
TNFd
STR-MICA
D6S2811
D6S510
HLA-B
231
216
244
153
139
126
186
110
178
8
1.8843
1.7729
1.7679
1.4323
2.2174
3.1648
2.1877
1.6667
1.3995
3.5476
0.1217
0.1852
0.0342
0.1804
0.0244
0.0069
0.0032
0.2119
0.2122
0.0238
0.9735
1.8518
0.3762
1.6234
0.3172
0.0552
0.016
4.6613
2.7583
0.6663
OR ⫽ odds ratio; pc values ⫽ p corrected for multiple tests.
frequency was the most discrepant between the two samples (7%).
However, this low statistic difference (OR ⫽ 3.55; 95% CI ⫽ 1.17–
10.72; p ⫽ 0.0238) was not maintained after multiple test correction (Table 3).
STR-MICA locus was in strong linkage disequilibrium with HLA-B
in both control (D’ ⫽ 0.7665; p ⬍ 0.00009) and patient (D’ ⫽ 0.7942;
p ⬍ 0.00009) samples. HLA-B*08 was associated with STR-MICA*183
in both samples (p ⬍ 0.002). Maximum likelihood haplotype
frequencies estimates indicate that the haplotype HLA-B*08:STRMICA*183 was the most frequent among patients (9%), but although also observed among controls (4%), it only ranked ninth
in frequency.
4. Discussion
The present study is a comprehensive mapping of the MHC
predisposing regions for ITP using microsatellite markers encompassing the three MHC regions and their main recombination
blocks. The ITP susceptibility region seemed to be located in the
surroundings of the STR-MICA locus, indicating class I centromeric
region, at the end of the ␤ block, as a primary predisposing region.
Previous studies have indicated STR-MICA associated with rheumatoid arthritis (weak association with MICA*04, secondary to HLADRB1*0405 association, and weak protective effect associated with
MICA*09 in Koreans [22]), psoriatic arthritis (disease development
associated with MICA*09, the same MICA*002 allele), and BehÈet’s
disease (associated with STR-MICA*06 in the absence of HLA-B*51
[15]). Those results indicate that the MICA gene may be involved in
autoimmune diseases etiology, including ITP. However, MICA maps
close to the HLA-B locus, which demonstrated an allele association
with ITP in previous studies [8,10]. Indeed, MICA and HLA-B loci
were in linkage disequilibrium in several populations studies [23–
27]. Moreover, some authors detected a strong association between
the STR-MICA A5.1 allele (STR-MICA*183 allele in our study), which
is associated with the MICA*008 extracellular allele [23], and the
HLA-B*08 allele [15,28,29]. However, the association of ITP with
HLA-B was weak and not retained after multiple test correction in
our study. This fact indicates that the primary association remains
with MICA, whereas HLA-B association seems to be secondary to
MICA, as supported by our linkage disequilibrium analysis.
Accordingly, some reported MHC class I and II associations with
ITP might be circumstantial as a result of linkage disequilibrium,
which partially explains the discrepancies found in previous studies. Indeed, single associations of HLA-B8, -B12 [8], and -A28 [9] with
ITP have been described; moreover, HLA-A3/B7/DR2 and HLA-A26/
B38/DR2 haplotypes, which are discrepant in their class I components, were also found to be associated with ITP [10]. On the other
hand, further studies have not confirmed any class I association
with ITP [30 –33], and different MHC class II alleles, HLADRB1*0410
and HLADRB1*0101, were associated with the disease in distinct
studies [5,6].
Because the association observed was restricted to STR-MICA, it
could not be explained by population stratification; otherwise, the
remaining microsatellites markers would have shown differences
in their allelic frequencies when comparing control and patient
samples. It is also important to highlight that the 51 ITP patients
studied, which seems to be a statistically small number, are an
appropriate sample because ITP is a fairly rare disease, as indicated
by sample sizes in other studies [5–7]. The prevalence rate of about
60 cases per million implies that our sample constitutes at least one
third of the predicted cases in the population surveyed in the
present study. In addition, highly stringent criteria were used to
select adult ITP carriers, preventing false-negative identification of
patients for autoantibodies.
Because disease association with microsatellite markers is possibly indirect, the MHC region close to STR-MICA could be the major
source of ITP susceptibility factors (i.e., the neighboring loci in this
region would be the best candidate genes for the disease). However, polymorphisms on STR-MICA may be immunologically relevant, because they have been associated with autoimmune dysfunctions like BehÈet’s disease and psoriatic arthritis [15,34]. In the
context of its immunological function, the MICA gene itself was
described as a ligand for the NKG2D/DAP10 stimulatory receptor
complex [35], expressed on natural killer cells, gd T cells and CD8⫹
ab T cells, triggering cytolysis mediated by NKG2D-bearing cells
[36] and also costimulating CD8⫹ ab T lymphocytes [37]. Although
MICA seems to be primarily involved with the cellular immune
response, it may be also implicated in the humoral response through
complex association between these two immune branches. For instance, once stimulated, NKG2D-bearing cells produce cytokines related to growth, proliferation, and differentiation of B cells into
plasmocytes, which play a central role in the production of antiplatelet IgM autoantibodies, characteristic of ITP etiopathogeny [3].
In short, the main conclusion of the present study indicates a
putative primary enrollment of genes very close to the MICA gene,
or even MICA itself, with ITP susceptibility.
Acknowledgments
This study was partially supported by the Brazilian National
Research Council (CNPq). We appreciate the collaboration of all
patients and thank the logistic and technical support personnel
provided by the Foundation Center of Hemotherapy and Hematology of Par and the Federal University of ParÂ.
References
[1] Chong BH, Ho SJ. Autoimmune thrombocytopenia. J Thromb Haemost 2005;3:
1763–72.
[2] Hopkins LM, John MD, Rico B, Rodney SV, Kenneth AS, John AG. MHC class
I-associated peptides identified from normal platelets and from individuals
with idiopathic thrombocytopenic purpura. Hum Immunol 2005;66:874 – 83.
[3] Cines DB, Blanchette S. Medical progress: immune thrombocytopenic purpura.
N Engl J Med 2002;346:995–1008.
[4] British Committee for Standards in Haematology: General Haematology Task
Force. Guidelines for the investigation and management of idiopathic thrombocytopenic purpura in adults, children and in pregnancy. Br J Haematol
2003;120:574 –96.
[5] Nomura S, Matsuzaki T, Ozaki Y, et al. Clinical significance of HLA-DRB1*0410
in Japanese patients with idiopathic thrombocytopenic purpura. Blood 1998;
91:3616 –22.
[6] Kuwana M, Kaburaki J, Pandey JP, et al. HLA class II alleles in Japanese patients
with immune thrombocytopenic purpura. Associations with anti-platelet glycoprotein autoantibodies and responses to splenectomy. Tissue Antigens
2000;56:337– 43.
[7] Satoh T, Pandey JP, Okazaki Y, et al. Single nucleotide polymorphisms of the
inflammatory cytokine genes in adults with chronic immune thrombocytopenic purpura. Br J Haematol 2004;124:796 – 801.
[8] Goebel KM, Hahn E, Havemann K. HLA matching in autoimmune thrombocytopenic purpura. Br J Haematol 1977;35:341–2.
[9] el-Khateeb MS, Awidi AS, Tarawneh MS, Abu-Khalaf M. HLA antigens, blood
groups and immunoglobulin levels in idiopathic thrombocytopenic purpura.
Acta Haematol 1986;76:110 – 4.
[10] Karpatkin S, Fotino M, Gibofsky A, Winchester RJ. Association of HLA-DRw2
with autoimmune thrombocytopenic purpura. J Clin Invest 1979;63:1085– 8.
M.H.T. Maia et al. / Human Immunology 70 (2009) 179-183
[11] Dawkins R, Leelayuwat C, Gaudieri S, et al. Genomics of the major histocompatibility complex: haplotypes, duplication, retroviruses and diseases. Immunol Rev 1999;167:275–304.
[12] Sambrook J, Fritch EF, Maniatis T. Molecular cloning: a laboratory manual, 2nd
ed. Cold Spring Harbor, Cold Spring Harbor Laboratory Press, 1989.
[13] Gourraud PA, Mano S, Barnetche T, Carrington M, Inoko H, Cambon-Thomsen
A. Integration of microsatellite characteristics in the MHC region: a literature
and sequence based analysis. Tissue Antigens 2004;64:543–55.
[14] Foissac A, Salhi M, Cambon-Thomsen A. Microsatellites in the HLA region:
1999 update. Tissue Antigens 2000;55:477–509.
[15] Mizuki N, Ota M, Kimura M, et al. Triplet repeat polymorphism in the transmembrane region of the MICA gene: a strong association of six repetitions with
BehÈet disease. Proc Natl Acad Sci USA 1997;94:1298 –303.
[16] Raymond M, Rousset F. GENEPOP (version 1.2): population genetics software
for exact tests and ecumenicism. J Heredity 1995;86:248 –9.
[17] Nei M. Molecular evolutionary genetics. New York, Columbia University Press,
1987.
[18] Sham PC, Curtis D. Monte Carlo tests for associations between diseases and
alleles at highly polymorphic loci. Ann Hum Genet 1995;59:97–105.
[19] Lancaster AK, Single RM, Solberg OD, Nelson MP, Thomson G. PyPop update—a
software pipeline for large-scale multilocus population genomics. Tissue Antigens 2007;69:192–7.
[20] Martin MP, Harding A, Chadwick R, et al. Characterization of 12 microsatellite
loci of the human MHC in a panel of reference cell lines. Immunogenetics
1998;47:131– 8.
[21] Malkki M, Single R, Carrington M, Thomson G, Petersdorf E. MHC microsatellite
diversity and linkage disequilibrium among common HLA-A, HLA-B, DRB1
haplotypes: implications for unrelated donor hematopoietic transplantation
and disease association studies. Tissue Antigens 2005;66:114 –24.
[22] Mok JW, Lee YJ, Kim JY, Lee EB, Song YW, Park MH, et al. Association of MICA
polymorphism with rheumatoid arthritis patients in Koreans. Hum Immunol
2003;64:1190 – 4.
[23] Tian W, Boggs DA, Ding W-Z, Chen D-F, Fraser PA. MICA genetic polymorphism
and linkage disequilibrium with HLA-B in 29 African-American families. Immunogenetics 2001;53:724 – 8.
[24] Zhang Y, Lazaro AM, Zou Y, Lavingia B, Moraes ME, Moraes JR, et al. MICA
polymorphism in South American Indians. Immunogenetics 2002;53:900 – 6.
183
[25] Tian W, Li L-X, Wang F, Luo Q-Z, Yan M-Y, Yu P, et al. MICA-STR, HLA-B
haplotypic diversity and linkage disequilibrium in the Hunan Han population
of southern China. Int J Immunogenet 2006;33:241–5.
[26] Gao X, Single RM, Karacki P, Marti D, O’Brien SJ, Carrington M. Diversity of MICA
and linkage disequilibrium with HLA-B in two North American populations.
Hum Immunol 2006;67:152– 8.
[27] Ribas F, Oliveira LA, Petzl-Erler ML, Bicalho MG. Major histocompatibility
complex class I chain-related gene A polymorphism and linkage disequilibrium with HLA-B alleles in Euro-Brazilians. Tissue Antigens 2008;72:532– 8.
[28] Petersdorf EW, Shuler KB, Longton GM, Spies T, Hansen JA. Population study of
allelic diversity in the human MHC class I-related MIC-A gene. Immunogenetics 1999;49:605–12.
[29] Bolognesi E, D’Alfonso S, Rolando V, Fasano ME, PraticÓ L, MomiglianoRichiardi P. MICA and MICB microsatellite alleles in HLA extended haplotypes.
Eur J Immunogenet 2001;28:523–30.
[30] Mueller-Eckhardt C, Mayr W, Lechner K, Mueller-Eckhardt G, Niessner H, et al.
HLA antigens in immunologic thrombocytopenic purpura (ITP). Scand J
Haematol 1979;23:348 –52.
[31] Mitchell EA, Birch HP, Woodfield DG. HLA and idiopathic thrombocytopenic
purpura. N Z Med J 1981;94:258 –9.
[32] Veenhoven WA, Sijpesteijn JA, van der Schans GS. HLA antigens in idiopathic
thrombocytopenic purpura. Acta Haematol 1979;62:153– 8.
[33] Mohanty D, Quadri MI, Das KC, Sengupta S. HLA and idiopathic thrombocytopenic purpura (ITP). Br J Haematol 1979;42:329 –30.
[34] Grubić Z, Perić P, EeÉuk-Jelicić E, et al. The MICA-A4 triplet repeats polymorphism in the transmembrane region confers additional risk for development of
psoriatic arthritis in the Croatian population. Eur J Immunogenet 2004;31:
93– 8.
[35] Wu J, Song Y, Bakker AB, et al. An activating immunoreceptor complex formed
by NKG2D and DAP10. Science 1999;285:730 –2.
[36] Bauer S, Groh V, Wu J, et al. Activation of NK cells and T cells by NKG2D, a
receptor for stress-inducible MICA. Science 1999;285:727–9.
[37] Groh V, Rhinehart R, Randolph-Habecker J, Topp, MS, Riddell SR, Spies T.
Costimulation of CD8alphabeta T cells by NKG2D via engagement by MIC
induced on virus-infected cells. Nat Immunol 2001;2:255– 60.