Molecular typing of HLA class II antigens in a São Paulo population

Molecular typing of HLA class II antigens in a São Paulo population

A.C. Goldberg¹, J.M. Chiarella¹, M.L.C. Marin¹, C. Rosales¹, D. Banic², M.A. Oliveira¹, H. Rodrigues¹, C.S. Viggiani¹ and J. Kalil¹

¹Laboratório de Imunologia de Transplantes, InCor, Faculdade de Medicina da Universidade São Paulo, SP, Brasil. Av. Eneas de Carvalho Aguiar, 500, 3⁰ andar, 005403-000 São Paulo, SP, Brasil. Fax: [phone]. E-mail: [e-mail] Send correspondence to A.C.G.

²Departamento de Imunologia, Fiocruz/Manguinhos, Rio de Janeiro, RJ, Brasil.

ABSTRACT

In the present paper we show data obtained from a normal population with a racially mixed profile typical of the city of São Paulo, State of São Paulo. Data were generated with polymerase chain reaction using sequence specific primers (PCR-SSP) for HLA-DRB and polymerase chain reaction followed by hybridization with sequence specific oligonucleotide probes (PCR-SSO) for HLA-DQA1 and HLA-DQB1 loci. HLA-DRB, DQA1, DQB1 and haplotype frequencies as well as common linkage disequilibria were found. This population was also shown to be in genetic equilibrium according to the Hardy-Weinberg law. HLA-DR typing of a normal sample from the city of Porto Velho, State of Rondonia, highlighted the importance of different sets of HLA profiles found in other regions of the country. This database provides essential information for screening studies of disease associations, forensic analyses and transplants.

INTRODUCTION

Human leukocyte antigens (HLA) constitute a group of major histocompatibility antigens directly involved in acquired immune responses (Colombani, 1993). The HLA molecule is responsible for peptide presentation to T lymphocytes and in this way triggers activation of different cells involved in immune response. The HLA molecule is in itself antigenic, eliciting strong alloreactive responses upon transplantation. Thus, correct definition of antigens in transplant donors and recipients adds significantly to the success of engraftment (Kerman, 1994).

The availability of more accurate molecular typing methods has allowed worldwide population data on HLA class II to be extensively reviewed and corrected. However, published data on HLA class II antigens are relatively rare in Brazil (Moraes and Moraes, 1996; Trachtenberg et al., 1988; Fernandez-Viña et al., 1991a,b; Moraes et al., 1993). Some studies have analyzed class II antigens by molecular typing. One analyzed DR and DQ alleles in Brazilian Amerindian samples (Fernandez-Viña et al., 1991a), while another compared HLA profiles of white and black individuals (Moraes et al., 1993). A third study established haplotypes in several populations including a group of 53 Latin Americans (Fernandez-Viña et al., 1991b). The Brazilian population in general and especially that of the city of São Paulo is unique. It has a high degree of admixture, and thus, we believe that it should be analyzed as one group. Defining the ethnic composition of these individuals is a difficult task. Many have different percentages of Caucasoid, Negroid and Amerindian ancestry. In addition, most European (predominantly Spanish, Portuguese and Italian) (Gorodezky et al., 1992), as well as Asiatic (Japanese, Korean) and African populations are represented in varying degrees. Strong internal migration from other regions of the country adds a further degree of complexity to this ethnical situation. It has, however, become increasingly clear that color does not define race (Goodman, 1997). Our results in some instances show precisely this phenomenon. In spite of this drawback, we believe that the data can provide vital information for those involved in clinical studies, transplantation and forensics.

One direct result of this situation is the difficulty in finding adequate sera for microlymphocytotoxicity typing of HLA class II antigens. Thus, we decided to examine the data obtained during the last 2 years in the course of routine polymerase chain reaction with sequence specific primers (PCR-SSP) typing of HLA-DRB1, B3, B4 and B5 (N = 308). Furthermore, 129 individuals also had their HLA-DQA1 and DQB1 alleles defined by polymerase chain reaction with sequence specific oligonucleotide probes (PCR-SSO). DRB1 typing was employed on samples from 80 normal individuals from Porto Velho, capital of the State of Rondonia in the Amazon Region. The data shown are an example of the care that should be taken when using data from other regions of Brazil.

MATERIAL AND METHODS

Samples

Blood samples were drawn from 308 individuals which either belonged to our local random cell panel (N = 99) or were kidney or blood marrow donors (N = 209). The sample consisted of 128 Caucasian, 21 mulatto, 7 black, 7 Asian and 146 non-color determined individuals. Samples (N = 80) from Porto Velho were also typed.

DNA was extracted alternatively by DTAB/CTAB or salting out methods (Bignon and Viña, 1995). In brief, buffy-coat from 4 ml of blood was mixed (about 1 ml) with 12% DTAB solution (12% DTAB, 2.25 M NaCl, 150 mM Tris, pH 8.6, 75 mM EDTA), vigorously mixed and incubated for 5 min at 68^oC. After the addition of 2 volumes of chloroform, samples were vigorously mixed and centrifuged for 2 min at 10000 g to separate into three phases. The upper phase was transferred to a new tube containing 2 volumes of 0.5% CTAB (0.5% CTAB, 0.04 M NaCl), DNA precipitated and washed further with 99.5% ethanol. When a salting method was used, blood was lysed with an ice-cold solution containing 10 mM Tris, pH 7.6, 5 mM MgCl₂ and 10 mM NaCl. After centrifugation, the process was repeated. Then, the white pellet was incubated in proteinase K buffer (0.375 M NaCl, 0.12 M EDTA, pH 8.0) containing 0.5% SDS and 2 mg of proteinase K for each 10 ml of blood, for 2 h at 56°C, at a final volume of 500 µl. The addition of 300 µl of 6 M NaCl and vigorous shaking were followed by centrifugation for 10 min at 1800 g to settle contaminating proteins. Supernatant containing DNA was mixed with 2 volumes 99.5% ethanol, precipitated, washed with 70% ethanol and centrifuged at 12000 g.

DRB1 typing

DR typing was performed by low resolution PCR-SSP according to Olerup and Zetterquist (1992).

DQA1 and DQB1 typing

PCR-SSO was performed using generic primers for exon-2 amplification according to protocols established by the 12th International Histocompatibility Workshop (Bignon and Viña, 1995). Generic DQA1 and DQB1 amplification was performed using the following primers:

DQA1 forward: 5' ATGGTGTAAACTTGTACCAGT 3'

reverse: 5' TTGGTAGCAGCGGTAGAGTTG 3'

DQB1 forward: 5' CATGTGCTACTTCACCAACGG 3'

reverse: 5' CTGGTAGTTGTGTCTGCACAC 3'

PCR conditions

Two hundred nanograms of DNA, 10 mM Tris-HCl, pH 8.4, 50 mM KCl, 2 mM MgCl₂, 0.001% gelatin, 0.2 mM dNTP, 20 pmol of each primer, 2 U Taq DNA polymerase (Cenbiot, RS, Brazil), H₂O up to 55 µl. Samples were cycled 30 times for 1 min at 94°C, 1 min at 55°C, 1 min at 72°C. Part of the PCR products (10%) were checked on 1.2% agarose gel electrophoresis with jX digested with HaeIII as a size marker. Two-5 µl/blots (total of 30-50 µl) of the amplified product were denatured in 160 µl NaOH 0.4 N/25 mM EDTA for 5 min at room temperature. Replicate filters were prepared by loading 30 µl of the denatured sample onto a Hybridot Manifold apparatus (BRL, Gaithersburg, MD, USA). The membranes were dried for 2 h at 80°C. For each oligonucleotide probe, 20 pmol was labeled with 40 µCi [g³²P] ATP using T₄ polynucleotide kinase. Blots were pre-hybridized (5 blots/bottle) with 10 ml of 6x SSPE, 5x Denhardt's solution, 0.5% SDS and 100 µg/ml of SSDNA for 4 h at 54°C and hybridized overnight with the added probe at the same temperature. Filters were washed twice in 2x SSC/0.1% SDS at room temperature, and once in TMAC solution (3 M TMAC, 50 mM Tris, 2 mM EDTA, 0.1% SDS, Denhardt 5x) for 45 min at 59°C. Before the exposition to X-ray film, membranes were washed in 2x SSC for a few minutes at room temperature. We reutilized each blot 2-3 times. To remove the probe, blots were washed 3 times in warm (95°C) TE 10-10 (10 mM Tris, pH 7.6/10 mM EDTA).

Statistics

Antigen and gene frequencies were calculated respectively by the formula: af = n/N, where n is the number of samples positive for the antigen, N is the total number of samples and gf = 1 - Ö(1 - af) (Baur and Danilovs, 1980). Haplotype frequencies were calculated by c² tests on 2 x 2 tables with help of the software "Étude haplotypique" contributed by the Laboratoire d'Histocompatibilité, Centre Régionale de Transfusion Sanguine (CRTS) at Rennes. The general formula for this calculation is: D(i,j) = p(i,j) - p(i)p(j), where p(i,j) is the haplotype frequency and p(i) and p(j) are the respective gene frequencies (Baur and Danilovs, 1980). Hardy-Weinberg equilibrium was tested calculating the sum of pairwise c² (Dausset and Pla, 1985).

RESULTS

Samples collected were partially obtained from our normal cell panel or from healthy transplant donors. The racial distribution of our samples was comparable to the official Brazilian Census of the São Paulo population, respectively: white, 79.9/69.9%; black or racially mixed, 17.1/27.9%; Oriental, 2.8/2.1%. However, 18.6% (DQA1/DQB1) and 47.2% (DRB1) of our samples came from individuals with racially undetermined origin. Thus, it is possible that the percentages would vary somewhat if these samples had been accounted for.

Table I shows the results obtained for DRB1 typing, as well as DRB3, 4 and 5 assignments for two populations. Some antigen frequencies are clearly different between the two populations, which highlights their ethnical differences. In the sample from Porto Velho, a big city in the Amazon Region, antigens of Amerindian origin such as DR16, DR4, DR14 and DR8 have increased frequencies. On the other hand, DR1 and DR17 have lower frequencies, reflecting the different admixture of races that constitute this segment of the Brazilian population. On the contrary, in the São Paulo population there is a marked presence of DR13 antigens.

Tables II

DRB1-DQB1 haplotype frequencies for the São Paulo population can be found in Table IV. This information can be useful for transplant donor matching.

Finally, Tables V and ^VI list statistically significant two-locus linkage disequilibria for DRB1-DQB1 and for DQA1-DQB1. Practically all common Caucasoid, Negroid and Amerindian linkages are present (see Discussion). We also tested our DR-typed population to see if it is in equilibrium according to the Hardy-Weinberg law. We obtained a c² value of 199 (P = 0.072) with 172 degrees of freedom, which is not significant and indicates that this population is in genetical equilibrium.

DISCUSSION

HLA typing with classical serological techniques is still widely used throughout the world. However, obtaining adequate results is fraught with difficulties due to strong cross reactivity between different antigens. In addition, in mixed populations such as the one found in São Paulo, traditional DR/DQB1 linkage disequilibria found especially in Caucasian populations can be further misleading for satisfactory assignment of class II antigens. For these reasons, reagents especially adapted to the population sample are essential for lymphocytotoxicity techniques. In our case, high crossreactivity in the DR52 group of antigens is especially troublesome due to the high frequency of DR13, DR14 and DR8 antigens, for which discriminating antisera are hard to obtain. A little under 35% of the São Paulo population has been typed DR13 or DR14, whereas 10% is positive for DR8.

Most of the common HLA-DR/DQB1 linkage disequilibria are found in the São Paulo population. Furthermore, in several instances rare haplotypes from different ethnical origins could be seen in our samples: DR11/DQB1*0602, DRB13/DQB1*0301, DRB16/DQB1*0301, DR7/DQB1* 0303, and DR9/DQB*0201, among others. These haplotypes were previously described (Moraes et al., 1993). Note that in the DR6 (DR13 and DR14) group of antigens, six different significant linkage disequilibria were found. Some more rare haplotypes such as DR15/DQB1*0601 and DRB1*1305/DQB1*0301 were present in this population at P < 0.05, but should be considered with care due to the very low number of individuals (n below 3) found for one of the antigen pairs. Linkage disequilibrium values in Table V are significant at P < 0.05 and should be reanalyzed with a larger sample size in order to validate the information. Conversely, linkage values of P < 0.001 were considered regardless of the number of samples found in this population.

The same general conclusions can be drawn upon examining the DQA1/DQB1 linkage data. Thus, typically Caucasian (DQA1*0102/DQB1*0602), African (DQA1*0102/DQB1*0501), Asian (DQA1*102/DQB1*601) or Amerindian (DQA1*0501/DQB1*0301) haplotypes can be found within the same group of DR51-linked antigens (DR15 and 16). Low number of positive samples led to the exclusion of the following linkage disequilibria from ^{Table VI}: DQA1*0103/DQB1*0608, DQA1*102/DQB1*601, DQA1*0101/DQB1*0503 and DQA1*03/DQB1*0303. Lastly, we could observe the presence of racially linked haplotypes in diversely classified individuals. An example would be an individual classified as white carrying the exclusively African haplotype HLA-DR18/DQB1*04. This occurred several times and strengthens our notion that race classification in Brazil is useless for genetical purposes. It should be noted that we applied the Hardy-Weinberg equation to our HLA-DR data (N = 308) to see if the population is at equilibrium. The result indicates that this population is in equilibrium and mixes randomly, in spite of social and ethnical barriers, and the strong internal migration currents that occur in Brazil.

RESUMO

Uma análise do perfil dos antígenos HLA de classe II numa amostra da população de São Paulo é descrita neste trabalho. Os dados foram obtidos através de técnicas de amplificação gênica utilizando-se iniciadores seqüências-específicos para HLA-DRB (PCR-SSP) ou amplificação gênica seguida de hibridação com oligonucleotídeos específicos (PCR-SSOP) para HLA-DQA e DQB. Foram calculadas as freqüências gênicas e as freqüências haplotípicas DRB-DQB e DQA-DQB e a população mostrou estar em equilíbrio genético de acordo com a lei de Hardy-Weinberg. Finalmente, comparamos também os dados obtidos com os da população de Porto Velho, Rondônia, salientando a importância da obtenção de dados regionais para os controles quando se estuda este complexo sistema genético.

ACKNOWLEDGMENTS

We are indebted to Dalia Ballas, M.S., for her invaluable help with statistical analyses. Research partially supported by PADCTII / SBIO (No. 620087/94.3), FAPESP (No. 93/3282-1) and CNPq. Publication supported by FAPESP.

(Received August 14, 1997)

• Baur, M.P. and Danilovs, J.A. (1980). Population analysis of HLA-A,B,C,DR and other genetic markers. In: Histocompatibility Testing (P.I. Terasaki, ed.). UCLA Tissue Typing Lab., Los Angeles, pp. 955-1210.
• Bignon, J.D. and Vińa, M.F. (1995). 12th IHWC HLA Class II Reference Protocol. In: Technical Handbook of the 12th International Histocompatibility Workshop (Fauchet, R. and Charron, D., eds.). HLA et Médecine, Paris.
• Colombani, J. (1993). HLA: Fonctions Immunitaires et Applications Médicales John Libbey Eurotext, Paris.
• Dausset, J. and Pla, M. (Eds.) (1985). Appendice. In: HLA: Complexe Majeur d'Histocompatibilité de l'Homme, Medecine-Sciences, Flammarion, Paris, pp. 116-117.
• Fernandez-Vińa, M., Moraes, J.R., Moraes, M.E., Miller, S. and Stastny, P. (1991a). HLA class II haplotypes in Amerindians and black North and South Americans. Tissue Antigens 38: 235-237.
• Fernandez-Vińa, M.A., Gao, X., Moraes, M.E., Moraes, J.R., Salatiel, I., Miller, S., Tsai, J., Sun, Y., An, J., Layrisse, Z., Gazit, E., Brautbar, C. and Stastny, P. (1991b). Alleles at four HLA class II loci determined by oligonucleotide hybridization and their associations in five ethnic groups. Immunogenetics 34: 299-312.
• Goodman, A.H. (1997). Bred in the bone? Sciences 37: 20-25.
• Gorodezky, C., Loon, J., Moliterno, R., Torres, E., Gerbase de Lima, M., Pelegrino, J., Ferreira, E., Alvarez, I., Carranza, J.M., Hernandez, O., de la Rosa, G., Debaz, H., Olivo, A., Moreno, M., Layrisse, Z. and Elias, M. (1992). HLA in some Latin American populations: Mexicans, Brazilians, Venezuelans and Uruguayans. In: Histocompatibility Testing 1991 (Tsuji, K., Aizawa, M. and Sasazuki, T., eds.). Vol. 1. Oxford University Press, New York, pp. 662-665.
• Kerman, R. (1994). To match or not to match. That's the question! In: Clinical Transplants 1993 (Terasaki, P.I. and Lecka, J.M., eds.). UCLA Tissue Typing Laboratory, Los Angeles, pp. 431.
• Moraes, J.R. and Moraes, M.E.H. (1996). Distribuiçăo dos alelos HLA na populaçăo brasileira. Hematologia Hemoterapia 1: 18-21.
• Moraes, M.E., Fernandez-Vińa, M., Salatiel, I., Tsai, S., Moraes, J.R. and Stastny, P. (1993). HLA class II DNA typing in two Brazilian populations. Tissue Antigens 41: 238- 242.
• Olerup, O. and Zetterquist, H. (1992). HLA-DR typing by PCR amplification with sequence-specific primers (PCR-SSP) in 2 hours: An alternative to serological DR typing in clinical practice including donor-recipient matching in cadaveric transplantation. Tissue Antigens, 39: 225-235.
• Trachtenberg, A., Jobim, L.F.J., Kraemer, M.E., Salzano, F.M., Moraes, M.E., Moraes, J.R.F., Gerbase de Lima, M., Arce-Gomez, B. and Ferreira, E. (1988). The HLA polymorphism in five Brazilian populations. Ann. Hum. Biol. 15: 213-221.

Publication Dates

• Publication in this collection
  23 Feb 1999
• Date of issue
  Sept 1998

History

• Received
  14 Aug 1997