Do FY Antigens Act As Minor Histocompatibility Antigensin the Graft-Versus-Host Disease Paradigm After Human

Leukocyte Antigen-Identical Sibling HematopoieticStem Cell Transplantation?

Mohamed Hichem Sellami,1 Manel Chaabane,1 Houda Kaabi,1 Lamia Torjemane,2

Saloua Ladeb,2 Tarek Ben Othmane,2 and Slama Hmida1

FY antigens are candidate minor histocompatibility antigens relevant to renal allograft rejection, but no datahave been reported about their role in graft-versus-host disease (GVHD) incidence after human leukocyteantigen (HLA)-identical siblings hematopoietic stem cell transplantation (HSCT). The aim of this study was toexamine the effect of donor/recipient disparity at FY antigens on the incidence of GVHD in Tunisian patientsreceiving an HLA-identical HSCT. This work enrolled 105 Tunisian pairs of recipients and their HLA-identicalsibling donors of HSCs. FY genotyping was performed with the polymerase chain reaction–sequence-specificprimer method and donor/recipient disparity for these antigens was analyzed at two levels: incompatibility andnonidentity. The case–control analyses showed no significant correlation between FY disparity and the incidenceof either acute or chronic GVHD. Sample size calculation showed that 572 cases and 1716 controls would benecessary to be able to detect a significant association with 80% power and two-sided type I error level of 5%(a = 0.05). The lack of association in the studied cohort may be explained by the low immunogenicity of FYantigens in HSCT context, compared with other antigens such as HA-1 and CD31.

Introduction

The ability of allogeneic stem cell transplantation(hematopoietic stem cell transplantation [HSCT]) to cure

several hematological malignancies has been widely rec-ognized ( Jenq and Van den Brink, 2010). However, thistherapy is often complicated by severe graft-versus-hostdisease (GVHD), which remains the major cause of mor-tality affecting about 50% of all patients even in humanleukocyte antigen (HLA)-identical siblings (Ferrara et al.,2009; Paczesny et al., 2010). The pathogenesis of GVHD ismediated by alloreactive T CD4+ and/or CD8+ cells di-rected against the recipient’s minor histocompatibility an-tigens (MiHAgs) that are not expressed in the donor (Sunet al., 2007; Ferrara et al., 2009; Paczesny et al., 2010). Mi-HAgs are a variety of endogenously synthesized polymor-phic products that can be recognized by alloreactive T cellsonly in the context of major histocompatibility complexmolecules (HLA) (Malarkannan et al., 2005; Hambach et al.,2007; Spencer et al., 2010). The importance of autosomal andsex-linked MiHAgs in the GVHD paradigm has been lar-gely described (Hambach et al., 2007; Sun et al., 2007; Toubai

et al., 2008; Ferrara et al., 2009; Socie and Blazar, 2009). Inthis regard, it has been demonstrated that disparity be-tween the recipient and his/her donor of HSCs for so-calledimmunodominant MiHAgs may be sufficient to causeGVHD, but further studies are needed to verify this corre-lation (Behar et al., 1996; Goulmy et al., 1996; Martin, 1997;Tseng et al., 1999; Gallardo et al., 2001; Grumet et al., 2001;Socie et al., 2001; Cavanagh et al., 2005; El-Chennawi et al.,2006; Stern et al., 2008; Markiewicz et al., 2009; Sellami et al.,2010). The FY antigens are candidate MiHAgs relevant torenal allograft rejection (Lerut et al., 2007), but no data havebeen reported about their role in GVHD incidence afterHLA-identical siblings HSCT. Hence, analyzing the donor/recipient FY disparity in association with GVHD occurrencemay be a useful way to examine their probable role asMiHAg.

FY antigens belong to the Duffy blood group system thatis of major interest in therapeutic medicine (Meny, 2010).This serologically determined system consists of two anti-thetic antigens, FY1 (Fya) and FY2 (Fyb), encoded by asingle FY gene that maps to chromosome 1q22–q23 (Lau-tenberger et al., 2000). This gene contains four major allelic

1The ‘‘Immunogenetic Applied to Cells Therapy’’ Research Unit, The Immunohaematology and HLA-Typing Department, National BloodTransfusion Centre of Tunis, Tunis, Tunisia.

2Department of Haematology, National Bone Marrow Transplantation Centre of Tunis, Tunis, Tunisia.

DNA AND CELL BIOLOGYVolume 31, Number 3, 2012ª Mary Ann Liebert, Inc.Pp. 331–336DOI: 10.1089/dna.2011.1321

331

forms: two codominant alleles FY*1 and FY*2 coding, re-spectively, for the FY1 and FY2 antigens; FY*0 is the FYsilent version and does not express FY antigens only at theerythroid level; and an FY*X allele producing a very weakFY2 antigen. The molecular bases of FY polymorphism havebeen well described (see, for example, Daniels, 2009).Interethnic analysis for FY alleles shows diversity whencomparing Caucasian and sub-Saharan populations. In fact,it has been reported that FY null phenotype (lack of FY1and FY2 antigens on RBCs) is very frequent in Africanpeople (*90%) and this has been explained by certainepidemiologic factors such as the natural selection withmalaria vivax, because Duffy molecule is the only RBCsreceptor for Plasmodium vivax (Kwiatkowski, 2005; Daniels,2009). In contrast, the FY-positive phenotype (expression ofFY1 and/or FY2 antigens on RBCs) is common amongCaucasians (Daniels, 2009). For Tunisians, FY allele fre-quencies were found (after having excluded FY*X, whichhad a frequency of 0.0174): FY*1 = 0.291 (expressed 0.260;silent 0.031); FY*2 = 0.709 (expressed 0.427; silent 0.282)(Sellami et al., 2008).

The Duffy molecule (gpD) is a transmembrane glycopro-tein of 35–43 kDa expressed in two forms a and b of 338 and336 amino acids, respectively. This molecule contains anNH2 extracellular domain containing 59 amino acids, whichseems to be very important at the histocompatibility level (acandidate MiHAg), especially in transplantation contexts(Tournamille et al., 2003; Lerut et al., 2007). In fact, thispeptide sequence contains the FY1/FY2, FY6, and a part ofthe FY3 antigen, which are linked to certain hemolytic dis-eases such as that affecting the newborn (HDN) (Tournamilleet al., 2003; Cotorruelo et al., 2009). Among others, Duffymolecule is expressed on RBCs, cerebellar neurons, epithelialcells of nonerythroid organs, and endothelial cells liningpostcapillary venules throughout the body, which representthe first site of contact with the donor alloreactive T cellsafter HSCT (Grimberg et al., 2007). Thus, we suspect thatDuffy molecule polymorphism can affect the occurrence ofthe GVHD reaction after HLA-identical HSCT by acting asMiHAg. Therefore, we decided to examine association in aTunisian cohort of patients receiving an HLA-identical sib-ling HSCT. For this purpose, we selected the FY*1/FY*2(rs12075) polymorphism coding for FY1/FY2 antigens formolecular analysis.

Materials and Methods

Informed consent was obtained from all participants and/or their families to participate in this study according to aprotocol approved by the ethical committee for scientific andmedical research of the National Blood Transfusion Centre ofTunis (Tunisia).

Patient selection

The cohort contains 105 Tunisian pairs of patients andtheir HLA-identical siblings. The patients have undergonethe HLA-matched HSCT in the National Bone MarrowTransplantation Centre of Tunis between January 2000 andMarch 2010. All patients received cyclosporine A and/ormethotrexate for GVHD prophylaxis and had either grades0–I/grades II–IV acute GVHD or chronic GVHD. For otherclinical characteristics, see Figure 1.

DNA samples

EDTA-treated blood samples were collected from all in-dividuals before HSCT. DNA was purified from peripheralblood leukocytes using the DNA blood minikit (Qiagen)according to the manufacturer’s instructions.

GVHD grading

The diagnosis and grading of acute and chronic GVHDwere defined according to previously published standardclinical criteria (Przepiorka et al., 1995).

DNA amplification using polymerase chainreaction–sequence-specific primer

Molecular amplification was obtained using specificprimers generating a 709-bp fragment for the FY alleles and a434-bp fragment for the internal control, which is a sequenceof the human growth hormone gene (HGH). The primersused in this work are those previously published (Mullighanet al., 1998). These primers were synthesized as oligonucle-otides from PROLIGO Primers & Probes.

Reaction mixtures and amplification conditions

Amplification was carried out in a final volume of 10mLcontaining 200mM of each dNTP, 0.2 mM of the detectionprimers (detecting the specific allele), 0.1 mM of the controlprimers (detecting the HGH sequence), and 1.5 mM MgCl2 in2 mL of 5X Green Buffer (Promega Corporation). Each poly-merase chain reaction (PCR) was performed using 150 ng ofgenomic DNA and 0.4 units of Taq polymerase (Go TaqTM

DNA Polymerase; Promega Corporation).The amplification protocol was 94�C for 5 min followed by

30 cycles of amplification consisting of denaturation at 94�Cfor 30 s, annealing at 60�C for 1 min, and elongation at 72�Cfor 30 s (Gene Amp PCR System 9600; Perkin Elmer). ThePCR products were visualized on agarose gels (1.5%) withUV illumination and photodocumented (UViTechTM).

FIG. 1. Clinical characteristics of the patient/donor pairsenrolled in the study (M, male; F, female; < or > 16, < or > 16years; Y, malignant; N, nonmalignant; TBI, conditioning withTBI; Other, other conditioning regimens.

332 SELLAMI ET AL.

Statistical analysis

Donor/recipient disparity at FY antigens is defined asfollows:

� Nonidentity: when the donor and the recipient are notcarriers of the same FY genotype.

� Incompatibility: when the donor is not compatible withthe recipient for the FY alleles.

Chi-square and Fisher’s exact tests were used to analyzethe association between both FY nonidentity/incompatibilityand either acute or chronic GVHD occurrence among thewhole cohort. The cohort was subdivided into subgroupsaccording to similarities of the peptide binding motif of theirHLA class I; and the association between both FY non-identity/incompatibility and either acute or chronic GVHDwas tested with univariate analyses. Subgroups for whichp-values were < 0.2 were reexamined in multivariate logisticregression models including major risk factors such asadulthood and conditioning.Odds ratios (OR) and p-valuesfor univariate and multivariate analyses were used to ana-lyze the correlation between risk factors and the probabilityof both acute and chronic GVHD. These analyses, modeledaccording to the default method and Hosmer-Lemeshowgoodness-of-fit, were performed with SPSS software (V 17.0).Positive correlation was considered only when p-value was< 0.05 (type I error level or a).

A sample size calculation was done using the Power andSample Size Program PS V3.0.43 ( January 2009) (Dupont andPlummer, 1998) to determine the sample size needed to de-tect a significant association between the donor/recipient FYdisparity and the GVHD incidence in the recipient of HSCs(i.e., to reject the null hypothesis H0). For this purpose, wemade the following assumptions: two-sided type I error level(a) of 0.05, a type II error level (b) of 0.2 (power of 80%), anda control/case ratio of 30.

Results

The molecular genotyping approach showed the presenceof all possible FY genotypes in the studied cohort. Clinically,the post-HSCT exams showed that all patients were evaluablefor acute GVHD versus only 86 cases for the chronic form.

General case–control analysis

In this part of the study, we preformed a general associ-ation analysis using donor/recipient FY disparity and inci-dence of GVHD after HSCT. This analysis showed nosignificant correlation between either FY nonidentity or in-compatibility and the incidence of acute and chronic GVHDamong recipients of HSCs (Fig. 2). Thus, we decided tosubdivide the cohort into subgroups according to similaritiesof the peptide binding motif of their HLA class I and reex-amine these associations separately in each group.

Subgroup case–control analyses

As there is a lack of a significant association in the generalanalysis, we suspected that subdividing the cohort into HLAclass I subgroups may be a useful way to reinforce the sta-tistical analysis. Subgroups are as follows: A2 (donor/recipient pairs carrier of the HLA-A*02 except A*0207,A*6802, and A*6901 alleles), other than A2 (pairs carrier of

HLA-A alleles other than of the first group), B7-like (pairscarrier of HLA-B7, -B35, -B51, -B53, -B54, -B55, -B56, -B67,-B78 alleles), B44-like (pairs carrier of HLA-B37, -B41, -B44,-B45, -B47, -B49, -B50, -B60, -B61 alleles), and other thanB7-like/B44-like (donor/recipient pairs carrier of HLA-B al-leles other than of B7-like/B44-like groups). In this model,we did not find any significant association (Figs. 3 and 4).Consequently, we decided to reanalyze only groups in whichwe observed a p-value of < 0.2 in a multivariate context in-cluding major risk factors such as adulthood and condi-tioning.

Multivariate analysis

This analysis considered only the relationship betweenFY1/FY2 identity and chronic GVHD incidence in the HLA-B7-like group ( p = 0.167, OR = 1.031). In this model, we ana-lyzed the combined effects of FY antigens identity, disease,donor/recipient sex mismatch, adulthood, and conditioning,which were described as potential risk factors (Fig. 5). As forthe other analyses, interactions between these factors werenot found to be significant in the GVHD paradigm.

FIG. 2. Analysis of association between both FY non-identity and incompatibility and incidence of either acute orchronic GVHD among patients enrolled in this study.GVHD, graft-versus-host disease.

FIG. 3. Univariate analysis of correlation between FYnonidentity and incompatibility and incidence of acuteGVHD among each HLA class I subgroup. HLA, humanleukocyte antigen.

FY ANTIGENS AND GVHD OCCURRENCE 333

Sample size calculation

We are planning a study of independent cases and con-trols with a control/case ratio of 30. Statistical data indicatethat the probability of exposure among controls is 0.27 andthe OR for disease in incompatible donor/recipient pairsrelative to compatible donor/recipient pairs is 1.3 (incom-patibility level) (Fig. 2). The sample size calculation foundthat a minimum of 572 cases and 1716 controls would benecessary to be able to detect a significant association with80% power and two-sided type I error level of 5% (a = 0.05).

Discussion

MiHAgs are allogeneic targets of T-cell–mediated GVHDeffects following HLA-identical siblings HSCT (Sun et al.,2007; Ferrara et al., 2009; Paczesny et al., 2010). Several im-munodominant MiHAgs have been identified, but defining astandard antigen panel remains a grand challenge for re-searchers. This difficulty is due to the high particularity of

these antigens that are a subset of endogenously synthesizedpolymorphic products encoded by autosomal and sex-linkedgenes and classified as non-HLA antigens (Goulmy, 1996).Thus, performing GVHD association studies using ubiqui-tous antigens such as those of some blood group systemsmay be a practical way to define more immunodominantsantigens. For these reasons, we decided to examine the effectof donor/recipient disparity in Duffy antigens (FY1/FY2) onthe incidence of GVHD after HLA-identical siblings HSCT,because these antigens are candidate MiHAgs relevant torenal allograft rejection (Lerut et al., 2007).

The general case–control analysis showed no significantcorrelation between donor/recipient disparity at FY antigensand the incidence of either acute or chronic GVHD amongrecipients of HSCs. Thus, we decided to subdivide the wholecohort into subgroups according to similarities of the peptidebinding motif of HLA class I to reinforce the statisticalanalysis but we did not find any significant results. The lackof association between FY disparity and GVHD incidencewas also observed when we performed a multivariate anal-ysis for the HLA-B7-like group that showed a p-value of< 0.2. In this model, we included more risk factors such asdonor/recipient sex mismatch and conditioning with TBI,but it appeared that the combined effects of these factorshave no influence on the pathogenesis of GVHD. Thus, itappears from these analyses that donor/recipient disparityat FY antigens is not critical in the GVHD paradigm in thestudied cohort. This finding may be explained by severallines of reasoning. First, we believe that the absence of rele-vant correlation is due to the low immunogenicity of the FYantigens compared with other antigens such as HA-1 andCD31. It has been reported that there is a difference at theimmunodominance levels between all described MiHAgs,because each one is a complex between an HLA ligand andan immunogen peptide and it is the combinatorial associa-tion that is recognized by the alloreactive T cells (Simpsonet al., 2002). Further, there are no data suggesting the in-volvement of FY antigens in T-cell–mediated immune re-sponses such as those in GVHD reaction. In fact,immunogenicity of FY antigens has been usually describedas potential cause of developing allo-antibodies, especiallyanti-FY1, and, therefore, humoral immunity (Meny, 2010).The contribution of humoral immunity in the GVHD para-digm is poorly defined, although that Miklos et al. reported acorrelation between antibody responses to H-Y MiHAg andchronic GVHD (2005). Thus, examining the anti-FY level insera of recipients who were mismatched with their donors ofHSCs at Duffy antigens may be useful.

Likewise, the use of a CSA/MTX-based prophylaxis pro-tocol by patients enrolled in this study may explain the lackof a significant association. This prophylaxis is used to pre-vent GVHD incidence after HSCT, and subsequently, the realnumber of patients who developed this outcome remainsrelatively ambiguous. Thus, we believe that the real effect ofFY disparity cannot be evaluated in subjects who receive thisprotocol. We need to investigate this effect in recipients whohad not received a GVHD prophylaxis. This procedure maybe feasible only in murine models, because the medical andethical duties require the physician to provide a prophylaxisprotocol to the patient.

To the best of our knowledge, this is the first work ex-amining the effect of donor/recipient FY disparity on GVHD

FIG. 4. Univariate analysis of correlation between FYnonidentity and incompatibility and incidence of chronicGVHD among each HLA class I subgroup.

FIG. 5. Multivariate logistic regression model of risk factorsfor chronic GVHD among HLA-B7-like positive patients.

334 SELLAMI ET AL.

incidence after HLA-identical siblings HSCT. The study re-vealed a lack of significant correlation, but further experi-ments are needed to confirm this finding. Further, we believethat it could be very important to investigate the effect ofFY disparity on other endpoints such as leukemia relapse,disease-free survival, and overall survival.

Acknowledgments

The authors thank the patients and their donors whoparticipated in this study. The authors are also very gratefulto the research and medical staff of the ‘‘National BloodTransfusion Center and National Bone Marrow Transplan-tation Centre of Tunis’’ for their valuable technical assistance.This work was supported by the Tunisian Ministry forHigher Education and Scientific Research (UR06SP04) andthe Tunisian Ministry of Public Health.

Disclosure Statement

No competing financial interests exist.

References

Behar, E., Chao, N.J., Hiraki, D.D., Krishnaswamy, S., Brown,B.W., Zehnder, J.L., and Grumet, F.C. (1996). Polymorphismsof adhesion molecule CD31 and its role in acute graft-versus-host disease. N Engl J Med 334, 286–291.

Cavanagh, G., Chapman, C.E., Carter, V., Dickinson, A.M., andMiddleton, P.G. (2005). Donor CD31 genotype impacts ontransplant complications after human leukocyte antigen-matched sibling allogeneic bone marrow transplantation.Transplantation 79, 602–605.

Cotorruelo, C., Biondi, C., Racca, L., Borras, S.G., and Racca, A.(2009). Duffy genotyping facilitates transfusion therapy. ClinExp Med 9, 249–251.

Daniels, G. (2009). The molecular genetics of blood grouppolymorphism. Hum Genet 126, 729–742.

Dupont, W.D., and Plummer, W.D. (1998). Power and samplesize calculations for studies involving linear regression. Con-trol Clin Trials 19, 589–601.

El-Chennawi, F.A., Kamel, H.A., Mosaad, Y.M., El-Sherbini,S.M., and El-Billey, N.A. (2006). Impact of CD31 mismatcheson the outcome of hematopoietic stem cell transplant of HLA-identical sibling. Haematology 11, 227–234.

Ferrara, J.L., Levine, J.E., Reddy, P., and Holler, E. (2009). Graft-versus-host disease. Lancet 373, 1550–1561.

Gallardo, D., Arostegui, J.I., Balas, A., Torres, A., Caballero, D.,Carreras, E., Brunet, S., Jimenez, A., Mataix, R., Serrano, D.,Vallejo, C., Sanz, G., Solano, C., Rodrıguez-Luaces, M., Marın,J., Baro, J., Sanz, C., Roman, J., Gonzalez, M., Martorell, J.,Sierra, J., Martın, C., De la Camara, R., and Granena, A. (2001).Disparity for the minor histocompatibility antigen HA-1 isassociated with an increased risk of acute graft-versus-hostdisease (GvHD) but it does not affect chronic GvHD incidence,disease-free survival or overall survival after allogeneic hu-man leukocyte antigen-identical sibling donor transplantation.Br J Haematol 114, 931–936.

Goulmy, E., Schipper, R., Pool, J., Blokland, E., Falkenburg, J.H.,Vossen, J., Gratwohla, A., Vogelsang, G.B., Van Houwelingen,H.C., and Van Rood, J.J. (1996). Mismatches of minor histo-compatibility antigens between HLA-identical donors and re-cipients and the development of graft-versus-host disease afterbone marrow transplantation. N Engl J Med 334, 281–285.

Goulmy, E. (1996). Human minor histocompatibility antigens.Curr Opin Immunol 8, 75–81.

Grimberg, B.T., Udomsangpetch, R., Xainli, J., McHenry, A.,Panichakul, T., Sattabongkot, J., Cui, L., Bockarie, M., Chitnis,C., Adams, J., Zimmerman, P.A., and King, C.L. (2007). Plas-modium vivax invasion of human erythrocytes inhibited byantibodies directed against the Duffy binding protein. PLoSMed 4, e337.

Grumet, F.C., Hiraki, D.D., Brown, B.W.M., Zehnder, J.L., Zacks,E.S., Draksharapu, A., Parnes, J., and Negrin, R.S. (2001).CD31 mismatching affects marrow transplantation outcome.Biol Blood Marrow Transplant 7, 503–512.

Hambach, L., Spierings, E., and Goulmy, E. (2007). Risk assess-ment in haematopoietic stem cell transplantation: minor his-tocompatibility antigens. Best Practice Res Clin Haematol 20,

171–187.Jenq, R.R., and Van den Brink, M.R. (2010). Allogeneic haema-

topoietic stem cell transplantation: individualized stem celland immune therapy of cancer. Nat Rev Cancer 10, 213–221.

Kwiatkowski, D.P. (2005). How malaria has affected the humangenome and what human genetics can teach us about malaria.Am J Hum Genet 77, 171–192.

Lautenberger, J.A., Stephens, J.C., O’Brien, S.J., and Smith, M.W.(2000). Significant admixture linkage disequilibrium across30 cM around the FY locus in African Americans. Am J HumGenet 66, 969–978.

Lerut, E., Van Damme, B., Noizat-Pirenne, F., Emonds, M.P.,Rouger, P., Vanrenterghem, Y., Pirenne, J., and Ansart-Pirenne, H. (2007). Duffy and Kidd blood group antigens:minor histocompatibility antigens involved in renal allograftrejection? Transfusion 47, 28–40.

Malarkannan, S., Regunathan, J., and Timler, A.M. (2005). Minorhistocompatibility antigens: molecular targets for im-munomodulation in tissue transplantation and tumor therapy.Clin Appl Immunol Rev 5, 95–109.

Markiewicz, M., Siekiera, U., Karolczyk, A., Szymszal, J., Helbig,G., Wojnar, J., Dzierzak-Mietla, M., and Kyrcz-Krzemien, S.(2009). Immunogenic disparities of 11 minor histocompatibil-ity antigens (mHAs) in HLA-matched unrelated allogeneichematopoietic SCT. Bone Marrow Transplant 43, 293–300.

Martin, P.J. (1997). How much benefit can be expected frommatching for minor antigens in allogeneic marrow transplan-tation? Bone Marrow Transplant 20, 97–100.

Meny, G.M. (2010). The Duffy blood group system: a review.Immunohematology 26, 51–56.

Miklos, D.B., Kim, H.T., Miller, K.H., Guo, L., Zorn, E., Lee, S.J.,Hochberg, E.P., Wu, C.J., Alyea, E.P., Cutler, C., Ho, V.,Soiffer, R.J., Antin, J.H., and Ritz, J. (2005). Antibody responsesto H-Y minor histocompatibility antigens correlate withchronic graft-versus-host disease and disease remission. Blood105, 2973–2978.

Mullighan, C.G., Marshall, S.E., Fanning, G.C., Briggs, D.C., andWelsh, K.I. (1998). Rapid haplotyping of mutations in theDuffy gene using the polymerase chain reaction and sequence-specific primers. Tissue Antigens 51, 195–199.

Paczesny, S., Hanauer, D., Sun, Y., and Reddy, P. (2010). Newperspectives on the biology of acute GVHD. Bone MarrowTransplant 45, 1–11.

Przepiorka, D., Weisdorf, D., Martin, P., Klingemann, H.G.,Beatty, P., Hows, J., and Thomas, E.D. (1995). Meeting report:consensus conference on acute GVHD grading. Bone MarrowTransplant 15, 825–828.

Sellami, M.H., Kaabi, H., Midouni, B., Dridi, A., Mojaat, N.,Boukef, M.K., and Hmida, S. (2008). Duffy blood group

FY ANTIGENS AND GVHD OCCURRENCE 335

system genotyping in an urban Tunisian population. AnnHum Biol 35, 406–415.

Sellami, M.H., Ladeb, S., Kaabi, H., Cherif, G., Torjemane, L.,Othman, T.B., and Hmida, S. (2010). Acute graft-vs.-host dis-ease correlates with the disparity for the PECAM-1 S536Npolymorphism only in the HLA-B44-like positive Tunisianrecipients of HSCs. Cell. Immunol. 265, 172–178.

Simpson, E., Scott, D., James, E., Lombardi, G., Cwynarski, K.,Dazzin, F., Millrain, M., and Dyson, P.J. (2002). Minor H an-tigens: genes and peptides. Transplant Immunol 10, 115–123.

Socie, G., and Blazar, B.R. (2009). Acute graft-versus-host dis-ease: from the bench to the bedside. Blood 114, 4327–4336.

Socie, G., Loiseau, P., Tamouza, R., Janin, A., Busson, M.,Gluckman, E., and Charron, D. (2001). Both genetic and clin-ical factors predict the development of graft-versus-host dis-ease after allogeneic hematopoietic stem cell transplantation.Transplantation 72, 699–706.

Spencer, C.T., Gilchuk, P., Dragovic, S.M., and Joyce, S. (2010).Minor histocompatibility antigens: presentation principles,recognition logic and the potential for a healing hand. CurrOpin Organ Transplant 15, 512–525.

Stern, M., Brand, R., De Witte, T., Sureda, A., Rocha, V., Pass-weg, J., Baldomero, H., Niederwieser, D., and Gratwohl, A.(2008). Female-versus-male alloreactivity as a model for minorhistocompatibility antigens in hematopoietic stem cell trans-plantation. Am J Transplant 8, 2149–2157.

Sun, Y., Tawara, I., Toubai, T., and Reddy, P. (2007). Patho-physiology of acute graft-vs-host disease: recent advances.Transl Res 150, 197–214.

Toubai, T., Sun, Y., and Reddy, P. (2008). GVHD pathophysi-ology: is acute different from chronic? Best Pract Res ClinHaematol 21, 101–117.

Tournamille, C., Filipe, A., Wasniowska, K., Gane, P., Lisowska,E., Cartron, J.P., Colin, Y., and Le Van Kim, C. (2003). Struc-ture-function analysis of the extracellular domains of theDuffy antigen/receptor for chemokines: characterization ofantibody and chemokine binding sites. Br J Haematol 122,

1014–1023.Tseng, L.H., Lin, M.T., Hansen, J.A., Gooley, T., Pei, J., Smith,

A.G., Martin, E.G., Petersdorf, E.W., and Martin, P.J. (1999).Correlation between disparity for the minor histocompatibilityantigen HA-1 and the development of acute graft-versus-hostdisease after allogeneic marrow transplantation. Blood 94,

2911–2914.

Address correspondence to:Mohamed Hichem Sellami, Ph.D.

The Immunohaematology and HLA-Typing DepartmentThe National Blood Transfusion Centre of Tunis

13 Rue Djbel Lakhdar—Bab SaadounTunis 1006

Tunisia

E-mail: sellamimh@laposte.net

Received for publication May 11, 2011; received in revisedform June 14, 2011; accepted June 14, 2011.

336 SELLAMI ET AL.