HLA Antigens in Brazilian Patients with Rheumatic Heart Disease
Gerbase-DeLima, HLA and Rheumatic Heart Disease
M. Gerbase-DeLima, J. Temin, L.C.N. Scala, D.V. Santos
From the Department of Pediatrics (M.G., J.T.) and the Department of Cardiology (D.V.S.), Escola Paulista de Medicina, Universidade Federal de São Paulo, and the Department of Medicine, Faculdade de Ciências Médicas, Universidade Federal de Mato Grosso (L.C.N.S.), Brazil.
Corresponding author:
Maria Gerbase de Lima, MD
Rua Pirandello, 709
04623-000, São Paulo, SP, Brazil
Telephone number: 55-11-5574-4426; 55-11-5574-0548
FAX: 55-11-570-1590
e-mail: ger.dped@epm.br
Background Evidence for genetic linkage between the HLA complex and susceptibility to rheumatic heart disease has been obtained by cosegregation analysis in families. However, the identification of the specific HLA gene(s) responsible for conferring this susceptibility, investigated in association studies in which the frequencies of HLA antigens are compared between patients and controls, has been elusive. The present study was designed to investigate possible HLA class II associations with rheumatic heart disease in the Brazilian population. Since genetic association may depend on the ethnic background, we included in our analysis both white and Mulatto Brazilian patients in an effort to identify susceptibility HLA markers that are either shared by or peculiar to different ethnic groups.
Methods and Results HLA-DR and DQ antigens were determined in 84 white and in 58 Mulatto patients with rheumatic heart disease and their frequencies were compared to those of ethnically matched controls. A highly significant association with HLA-DR7 was detected in the Mulatto patients, whereas a trend for an association with HLA-DR1 was observed in the white patients.
Conclusions Our results are interesting because they show clear differences in HLA associations between two different Brazilian ethnic groups (DR7 in Mullato subjects and DR1 in white subjects) living in the same geographical, thus suggesting that the different HLA associations with rheumatic heart disease reported in diverse populations are due to genetic rather than to environmental factors.
Condensed Abstract
In order to investigate HLA class II associations with rheumatic heart disease in the Brazilian population, HLA-DR and DQ antigens were determined in 84 white and in 58 Mulatto patients with rheumatic heart disease. A highly significant association with HLA-DR7 was detected in the Mulatto group of patients, whereas an association with HLA-DR1 was observed in the white patients. Thus, our resuts have shown different HLA associations between two ethnic groups living in the same area. Therefore, we suggest that the different HLA associations with rheumatic heart disease that have been reported in the literature are due to genetic rather than to environmental factors.
Key words: · Rheumatic heart disease · genetics · histocompatibility antigens· MHC·
The genes of the HLA complex stand as attractive candidates to play a role in the genetic susceptibility to rheumatic heart disease (RHD) in view of their extremely polymorphic character and the pivotal role of the HLA molecules (antigens) in the presentation of peptide fragments to T cell receptors. Previous studies concerning the analysis of the inheritance of the HLA haplotypes in families with more than one individual with rheumatic fever (RF) provided evidence for the presence of a RF susceptibility gene located within or very near the HLA complex1,2. Several studies in which the frequencies of HLA antigens were compared between patients and controls showed some associations with HLA class I antigens (HLA-A,B) which, however, were weak and inconsistent3-5. Concerning HLA class II antigens (HLA-DR, DQ), the data are more impressive, although there is poor agreement among studies regarding the associated HLA antigen(s)6,7. Association with HLA-DR4 was reported in Caucasians from the United States8,9 and Saudi Arabia10 association with DR1 was reported in black patients from Martinique11 and South Africa12, and an association with HLA-DR2 was reported in black Americans8. HLA-DQ2 was associated in Indian subjects13, and DR3 and DR7 were associated in Turkish patients14. A study on Brazilian patients of admixed genetic background disclosed an association with HLA-DR7 and DR5315. The lack of consistency of the associations found in the different studies could be attributed to environmental and/or to genetic background differences.
The present report refers to a study on HLA and RHD in the Brazilian population in which the patients were subdivided according to their main genetic background, with one group consisting of white patients and the other of Mulatto patients. Since all subjects came from practically the same geographical area, we thought that this kind of study could shed some light on the issue of whether the differences among the HLA associations found in different populations could be mainly accounted for by environmental, like regional differences in streptococcal outbreaks, or by genetic factors.
Methods
Patients
The study was conducted on 84 white and 58 Mulatto Brazilian patients with rheumatic heart disease. The diagnosis of RHD was based on medical history, physical examination, available previous medical records, chest x-ray, ECG, and Doppler echocardiogram. The histological examination of the valves of patients who underwent cardiac surgery was also taken into consideration. All patients were selected by one of us (L.C.N.S.) at the outpatient clinic of the Cardiology Department of Escola Paulista de Medicina, Universidade Federal de São Paulo. The protocol was approved by the Ethical Committee of Escola Paulista de Medicina and informed consent to participate in the study was obtained from all subjects.
HLA Typing
HLA-DR and DQ antigens were determined by the microlymphocytotoxicity technique16 and commercial anti-sera (Biotest Diagnostic, Frankfurt, Germany; One Lambda Inc., Los Angeles, USA) were used to define 12 HLA-DR (DR1-DR10, and DR52, DR53) and three DQ (DQ1-DQ3) specificities.
Statistical Analysis
The HLA antigenic frequencies observed in the patients were compared to those of local ethnically matched control groups, by the Fisher’s exact test. The P values were corrected (Pc) for the number of antigens tested17, i.e., the P values were multiplied by 15. In cases of significant differences after correction (Pc < 0.05), the strength of the association was assessed using the odds ratio (OR) as an approximation to estimate the relative risk (RR), and the etiologic fraction (EF) was calculated according to the following formula: EF = [(RR - 1)/RR] x frequency of the trait in patients17.
Results
The phenotypic frequencies of HLA-DR and DQ antigens in the two patient groups, as well as in their respective controls, are shown in Table 1. The increased frequency of HLA-DR7 in the Mulatto patients was the only difference that continued to be significant after correction for the number of antigens tested. In this group of patients the frequencies of DR53 and DQ2 were also increased, with respective P values, not corrected for the number of comparisons, of 0.03 and 0.027. Among the white patients, the frequency of DR1 was increased (30.9 vs 16.6; P = 0.005; Pc = 0.07) and of DQ1 (65.8 vs 50.7; P = 0.034; Pc > 0.05).
Discussion
The objective of the present study was to investigate possible HLA-DR and DQ associations with rheumatic heart disease. Since genetic association may depend on the ethnic group under study, we included in our analysis both white and Mulatto patients in an effort to identify genetic components that are either shared by or peculiar to different ethnic groups.
We found a significant increase in HLA-DR7 frequency in the Mulatto population, with this antigen being present in 38.9% of the patients and in 11.7% of the controls (P = 0.0006, Pc = 0.009). The relative risk associated with DR7 was 4.7, meaning that individuals with this marker are about five times more susceptible to rheumatic heart disease than individuals from the general population who do not carry DR7. Compared with other HLA-associated diseases, this figure is small. The ethiologic fraction, a measure that is supposed to reflect how much of a disease "is due to" the presence of a given character17, was found to be 0.30. In this same group of patients, the frequencies of HLA-DR53 and DQ2 were also elevated, which is easily explained considering that the gene coding for DR7 is almost always associated at the chromosome level (i.e., in very strong linkage disequilibrium) with the genes that code for DR53 and DQ2.
The serological DR7 specificity corresponds to only one allele, i.e., no further DR7 polymorphisms have been detected thus far at the DNA level, the same being true for HLA-DR5318. Therefore, we may infer that the associations observed here in the Mulatto population correspond to alleles DRB1*0701 and DRB4*01. The DQ2 antigen may correspond to three different alleles18, but considering the population frequencies, as well as the linkage disequilibrium with DR719, the most probable associated allele appears to be DQB1*0201.
An association of DR7 and DR53 with rheumatic heart disease has been previously reported in Brazilian patients of admixed genetic background15 and we believe that the finding of the same association in the present Mulatto patients is of great interest.
In contrast, no indication of an association with HLA-DR7 was observed in the white patients. In this group we found an increase in DR1 frequency, with this antigen being present in 30.9% of the patients and in 16.6% of the controls (P = 0.005). Although the corrected P value did not reach significance at the usual 0.05 level, it was close to it (0.07). In addition, we observed a less important increase in the DQ1 frequency and we interpret this finding as a consequence of the known linkage disequilibrium between DR1 and DQ1.
The possible association of RHD and DR1 in the white population has no parallel in the literature concerning studies on Caucasian populations, but is in agreement with the findings of some authors that have studied black populations11,12.
It is interesting to note that stronger associations with HLA-DR than with HLA-DQ antigens were found in both groups of patients, which may imply that the susceptibility to rheumatic heart disease is more closely related to DR than to DQ antigens, as has been claimed by other authors7,15.
In conclusion, the present study on Brazilian subjects from practically the same geographical area has clearly shown different associations between HLA antigens and rheumatic heart disease, depending on the ethnic background of the subjects. Therefore, we believe that the lack of consistence among studies conducted on different populations regarding the HLA associations, is better explained by genetic than by environmental factors.
Supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico and Fundação de Amparo à Pesquisa do Estado de São Paulo.
References