Allelic polymorphisms of human platelets-specific alloantigens in South Tunisian population

Abstract

Objectives

Human platelet-specific alloantigens (HPA) are polymorphic epitopes which vary among ethnic groups.

Background

In Tunisia, HPA frequencies were determined in North and centre; however, the pattern of HPA in South Tunisian population is not been studied yet. The aim of this work was to determine allelic frequencies of HPA-1, -3, and -5 systems in south Tunisian population, in order to estimate the risk of anti-platelet allo-immunization and to create a register of HPA-typed blood donors.

Methods

Our study concerned 212 unrelated healthy, regular blood donors from southern Tunisia. Allelic polymorphisms of each system were determined using a polymerase chain reaction with sequence-specific primers.

Results

Genotype frequencies a/a, a/b, and b/b were, respectively, 0.670, 0.288, and 0.042 for HPA-1 system, 0.430, 0.462, and 0.108 for HPA-3 system, and 0.750, 0.241, and 0.009 for HPA-5 system. The allele frequencies were 0.814 and 0.186 for HPA-1a and -1b alleles; 0.660 and 0.340 for HPA-3a and -3b alleles and 0.870, and 0.130 for HPA-5a and -5b alleles.

Discussion

The reported frequencies are more similar to those of Caucasians than those of north Tunisian population.

Keywords:

• Human platelets antigens
• PCR-SSP
• Allelic polymorphism
• Gene frequency
• Tunisian population

Introduction

Human platelet-specific alloantigens (HPA) are polymorphic epitopes of platelet membrane glycoproteins (GP).¹ To date, 24 platelet-specific alloantigens have been defined by immune sera, of which 12 are grouped into six bi-allelic systems (HPA-1, -2, -3, -4, -5, and -15). The molecular basis of 22 of the 24 serologically defined antigens has been identified. For 21 platelet allo-antigens, the difference between self and non-self is distinguished by a single amino acid substitution, caused by a single-nucleotide polymorphism (SNP) in the gene encoding the relevant GP membrane.²

These epitopes are the target of platelet alloantibodies which can appear during pregnancy or after blood transfusion.¹ HPA allo-immunization is commonly responsible of three important clinical manifestations: Neonatal allo-immune thrombocytopenia (NAIT), post-transfusional purpura (PTP), and platelet transfusion refractoriness (PTR).^3,4

The relative importance of each HPA system in anti-platelet allo-immunization varies among ethnic groups.^5–7 Thus, the study of HPA polymorphism in different populations presents several interests. The first one is epidemiologically important in clinical practice and particularly the prediction of alloimmune disease.⁸ The second one gives support allo-immunized patients by the determination of HPA phenotypes and anti-HPA antibodies in order to allow the diagnosis, establish a prognosis, and to provide compatible platelets when needed.^9,10 Finally, these studies present a genetic anthropological interest.⁸

Serological HPA phenotyping is very limited while platelet genotyping by molecular approach is widely recommended. Several techniques for SNP typing have been described for platelet genotyping. Polymerase chain reaction (PCR) amplification with sequence-specific primers (PCR-SSP) is currently used.⁷

The aim of this work was to determinate the allelic frequencies of HPA-1, -3, and -5 systems, in order to estimate the risk of anti-platelet allo-immunization and to create a register of HPA-typed blood donors in the south of Tunisia.

Materials and methods

Blood donors, samples, and gene amplification

This study concerned 212 blood donors (148 men and 64 women). All of them are unrelated regular healthy volunteers, from southern Tunisia (Sfax City: n =172, Delegations of Sfax and South: n = 40). They were divided into 147 whole-blood donors and 65 platelets donors. Genomic DNA was extracted by the salting-out method,¹¹ from venous blood samples of 10 ml each collected on EDTA.

HPA-1, -3, and -5 systems were genotyped by PCR-SSP. The sequences of primers used here are according to Klüter et al.⁷ A pair of primers amplifying a conserved region of the gene for C-reactive protein (CRP) was included in each tube, as internal control (Table 1).

Table 1. Sequences and length of designed primers for HPA-1, -3, -5 and the two internal positive control primers

Primer code	Primer sequences	Length primers	PCR products length (bp)
HPA-1a-I	ACTTACAGGCCCTGCCTCT	19	189
HPA-1b-I	ACTTACAGGCCCTGCCTCC	19
HPA-1-II (common)	GTGCAATCCTCTGGGGACT	19
HPA-3a-I	GGGGGAGGGGCTGGGGA	17	293
HPA-3b-I	GGGGGAGGGGCTGGGGC	17
HPA-3-II (common)	GGCCCTGGGACTGTGAATG	19
HPA-5a-I	AGGAAGAGTCTACCTGTTTACTATCAAAG	29	256
HPA-5b-I	AGGAAGAGTCTACCTGTTTACTATCAAAA	29
HPA-5-II (common)	CTCTCATGGAAAATGGCAGTA	21
CRP-I	CCAGCCTCTCTCATGCTTTTGGCCAGACAG	30	440
CRP-II	GGGTCGAGGACAGTTCCGTGTAGAAGTGGA	30

I: forward primer; II: reverse primer

PCR was performed as described previously by Klüter et al.⁷ using a Gene Amp PCR System 9700, Applied Biosystem Thermal Cycler (Singapore), in a volume of 20 µl. Positive and negative controls were included in each series.

PCR products were visualized by UV trans-illumination after electrophoresis on agarose gel at 1–2% and staining with ethidium bromide.

Statistical analysis

Genotype frequencies were determined by direct counting. Allele frequencies were calculated from genotype frequencies. The confidence interval was set at 5%. The validity of Hardy–Weinberg equilibrium for each of the HPA system was tested by χ² test.¹² Comparisons of allele frequencies between different populations were assessed by the χ² test (reduced gap). The significance level was chosen to be 5%.

Results

In our study, PCR conditions were set different for HPA-1b, -3b, -5a, and -5b alleles. In fact, the alleles of the same system have different hybridization temperatures. Our work was therefore organized in several series of blood donors for each allele.

We have chosen nine HPA-specific primers and two CRP primers (internal positive control) for HPA-1, -3, and -5 systems genotyping. We used also DNAs with known HPA genotypes. In Fig. 1, we give an example of the results obtained by the PCR-SSP method showing that primers specifically amplify the desired platelet HPA-1a-related DNA and that the two alloantigen-specific primers clearly distinguish between the two alleles. The 440 bp amplification product of CRP control primer was present in all lanes. Similar PCR amplification patterns were obtained with primers related to the other platelet-specific alloantigens (HPA-3 and -5).

Figure 1. Examples of amplification in HPA-1a system. The upper band corresponds to the amplification product of the CRP gene (internal control of 440 bp). The lower band corresponds to the amplification product of HPA-1a gene (189 bp) situated between 123 and 246 bp bands of molecular weight markers.Lane1: negative control; lanes 2, 3, 4, 6, and 7: individuals with HPA-1a allele; lane 5: individual without HPA-1a allele; lane 8: molecular weight marker of 123 bp.

Using this procedure, we typed 212 Southern Tunisian blood donors for HPA-1, -3, and -5 systems. Tables 2 and 3 summarize genotype and allele frequencies. There is no significant deviation from the Hardy–Weinberg equilibrium in any of the studied systems in this population (respectively, χ² 0.55; 0.2 and 0.9).

Table 2. Genotype frequencies of HPA-1, -3, and -5

Genotypes	HPA-1	HPA-3	HPA-5
a/a	0.670	0.430	0.750
a/b	0.288	0.462	0.241
b/b	0.042	0.108	0.009

Table 3. Frequencies of HPA alleles in different ethnic groups

Country	Ethnic groups	Effective	HPA-1a	HPA-1b	HPA-3a	HPA-3b	HPA-5a	HPA-5b	Method
Chine^Citation16	Chinese	1000	0.994*	0.006^†	0.5945	0.4055	0.9860*	0.014^†	PCR-SSP
Japon^Citation30	Blood donors Japanese	331	0.991*	0.009^†	0.718	0.282	0.973*	0.027^†
Brazil^Citation20	Caucasian	100	0.925*	0.075^†	0.600	0.400	0.920	0.080	PCR-RFLP
Pakistan^Citation19	Major ethnic groups and castes of Pakistani population (Pathans and Sindhis)	593	0.885*	0.115^†	0.69	0.31	0.9	0.1	PCR-SSP
Austria^Citation21	Caucasoid of mixed background	911–906–931	0.852	0.148	0.612	0.388	0.892	0.108	PCR-RFLP
France^Citation22	Caucasian random platelet donors	6192	0.848	0.152	0.62	0.38	0.874	0.126	PCR SSP
Germany^Citation18	Random blood donor, estimated 99% Caucasian	1583–1562–1643	0.84	0.16	0.60	0.40	0.92*	0.08^†	PCR SSP
Germany^Citation7	Caucasians from Northen Germany	573	0.839	0.161	0.586	0.414	0.917*	0.083^†	PCR SSP
Croatia^Citation18	Caucasiens	430	0.83	0.17	0.64	0.36	0.89	0.11	PCR-SSP
Algeria^Citation21	–		0.827	0.173	–	–	0.837	0.163
Spain^Citation18	Spanish	727–662–454	0.81	0.19	0.65	0.35	0.88	0.12
Marocco^Citation15	Berber	107–104–112	0.748	0.252	0.682	0.318	0.861	0.139	PCR-SSP
Tunisia	Blood donors (Sfax)	212	0.814	0.186	0.660	0.340	0.870	0.130	PCR-SSP
Tunisia^Citation23	Blood donors (Tunis)	78–90–93	0.75	0.25	0.694	0.306	0.779^†	0.221*	PCR-RFLP
Tunisia^Citation26,Citation27	Blood donors (Tunis)	100	0.73	0.27	0.765	0.235	0.665^†	0.335*	PCR-SSP
Tunisia^Citation24	Central Tunisia (Monastir)	316	0.665^†	0.335*					PCR-SSP
Tunisia^Citation25	Sousse	444	0.59^†	0.41*	0.75*	0.25^†			PCR-SSP

RFLP, restriction-fragment-length polymorphism

*Frequency significantly higher than Southern Tunisia.

^†Frequency significantly lower than Southern Tunisia.

Table 3 showed also a comparison between our results and those observed in other studied populations.

Discussion

The present study is the first to investigate allele frequency of the HPA-1, -3, and -5 systems in Southern Tunisian population and serves as an outline for future clinical research associated with platelet disorders in this group. A HPA-typed platelet of 212 volunteers and regular blood donor registry was established. In our blood donors, a and b alleles frequencies of the three systems HPA-1, -3, and -5 are similar to those characterizing Caucasian populations.

PCR-SSP, adapted in our laboratory, is the most used technique for HPA genotyping.⁷ It is a rapid, simple and non-laborious technique. PCR products are visualized directly after gel migration. The required reagents and equipments are standardized, which make PCR-SSP relatively inexpensive.¹³ In routine, the cost of HPA genotyping may be reduced without compromising the resolution. This is achieved by limiting the testing to the most involved systems in clinical pathology in a given population.¹⁴ For this reason, our choice is limited to three systems HPA-1, -3, and -5: the most involved among Caucasians.⁷ Genotyping results of southern Tunisia population are shown in Table 3. The comparison between our results and the literature showed that

• - HPA-1a allele frequency varies from north to south in Tunisia. In the centre, HPA-1a frequency was the lowest value observed in the literature (0.59 and 0.66). These frequencies are significantly lower than those found in our study.^24,25 When compared to north Tunisia, our HPA-1a frequency showed no significant difference.²³ Alleles distribution could be explained by predominant Berber influence in North and South, consanguineous marriages and technical considerations. In fact, in our study, blood donors which are negative for HPA-1a allele, were systematically checked and sometimes proved positive, which was not reported in the methodology of other studies.^23,26,27

• - Compared to other countries, the prevalence of the HPA-1a allele among southern Tunisian (0.814) was lower than that reported for Asians (Chinese, Japanese, and Pakistani)^16,19,29 and Brazilians.²⁰ Nevertheless, the prevalence of this allele in south Tunisia was similar to those reported in other Caucasian populations.^15,22

• - In southern Tunisia, the HPA-3 polymorphism system shows a similar prevalence in comparison with all other populations so far studied.

• - When compared to northern Tunisia, HPA-5a frequency in our blood donors was significantly higher.²³ Thus, the high frequency of 5b allele in north Tunisia could be explained by a large introduction of this allele through black slaves. The same conclusion was reported in studies on sickle cell anemia with the import of the Benin gene into Tunisia from black Africa.²⁸

• - HPA-5 allele frequencies analysis in southern Tunisia revealed a lower 5a allele frequency to those reported for Chinese¹⁶ and Japanese.²⁹ No statistically significant difference was found when compared to Berbers of Morocco, French, and Brazilian Caucasians.^15,20,22

These findings could be explained by the nature and the diversity of ethnic groups in Tunisian population. Indeed, Berbers, the original inhabitants of Tunisia and western part of North Africa, have been successively colonized and brewed with Phoenicians, Canaan (territory extended from Syria to Egypt through Lebanon and Palestine), Romans, Byzantines, Andalusians, Jews, Black slaves from trade between Sudan and Europe, and Turks.²³

Conclusion

The study of allelic polymorphism of HPA is very important, not only for anthropological and genetic reasons, but also to better predict the risk for allo-immunization for HPAs among distinct ethnic groups. The relative importance of each system HPA in anti-platelet allo-immunization varies among ethnic groups. Based on what has already been reported among Caucasians, our choice was limited to three systems HPA-1, -3, and -5.¹⁷ Allele frequencies of the three systems in south Tunisia were similar to those observed in Caucasians populations. The risk of immunization would join therefore that of Caucasians but seems to be less important than Northern Tunisia.

This study allowed us to establish a HPA-typed platelet registry of 212 volunteers and regular blood donor. This registry could help in contacting blood donors of rare and negative phenotypes who are involved in platelets donation. The use of typed platelets is highly beneficial and even indispensable in certain diagnostic situations (research and identification of specific anti-platelet allo-antibodies) and/or therapeutic (NAIT).

Acknowledgments

This research was supported by CRTS (Centre Régional de Transfusion Sanguine de Sfax) and CBS (Centre de Biotechnologie de Sfax), Tunisia.