Rheumatic fever is expressed as an inflammatory reaction that involves many organs, primarily the heart, the joints, and the central nervous system. The clinical manifestation of acute rheumatic fever (ARF) represents an abnormal host response following group A streptococcal infection of the tonsillopharynx. The major importance of ARF is its ability to cause fibrosis of heart valves, leading to crippling chronic rheumatic heart disease (RHD) [8].

Studies on individual host response together with the observation of familial incidence of the disease suggest that genetic factors play a role in susceptibility to rheumatic fever. Genetic environmental interaction, HLA, B cell alloantigens, and blood group associations have been demonstrated in studies of rheumatic fever among Egyptians [1, 11, 14, 21, 34].

Cytokines appear to play a critical role in triggering immunologic and inflammatory reactions in rheumatic fever. It was reported that blood mononuclear cell cultures from rheumatic children produced more tumor necrosis factor-α (TNF-α) than those from controls [26]. In addition, significantly elevated amounts of interleukin-1 (IL-1) have been found in ARF and active RHD patients at all time points up to 48 weeks [27]. Moreover, IL-6 and TNF-α are considered well-known inducers of acute phase reactants and show significant correlation with the C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) [40]. Others have suggested that estimation of IL-1α in carditis and IL-6 in arthritis may be helpful as minor criteria for diagnosis and follow-up of rheumatic activity, and advised therapy for rheumatic carditis with anticytokines such as anti-IL-1α and anti-TNF-α immediately after diagnosis to prevent or reduce valvular damage [12].

The implication of a heritable genetic basis of cytokine production could account for interindividual differences in immune responsiveness [24, 38]. The basis for these heritable differences is not known; however, polymorphism is one of the most reliable factors [3]. Two gene polymorphisms at positions −308 and −238 have been described within the promoter region of the TNF-α gene on chromosome 6, which are associated with differences in susceptibility or resistance to different diseases [22]. Several single base pair substitutions spanning the promoter region of IL-10 on chromosome 1 have also been identified, including positions -1082 G/A, -819 C/T, and -592 C/A, with a growing body of evidence suggesting a role for these polymorphisms in the protection, induction, and/or maintenance of inflammatory conditions [7, 32]. Similarly, a polymorphism in the 5′ flanking region of the IL-6 gene on chromosome 7 at position -174 has been reported. Specifically, subjects with the G allele showed higher plasma IL-6 levels than did carriers of the C allele [9]. The IL-1 cluster on human chromosome 2q12–2q14 harbors various promising candidate genes for inflammatory diseases. Of note, the IL-1Ra gene (IL-1RN) has a variable number tandem repeats (VNTRs) in intron 2, with up to five variants depending on the number of repeats of the 86−base pair (bp) fragment. The most common alleles have been termed allele 1 (A1, four repeats) and allele 2 (A2, two repeats), in addition to three other alleles [35, 36]. Allele 2 has been shown to be associated with increased production of IL-1Ra in vitro [17].

Taking into consideration that cytokine gene polymorphisms are population specific, we tested the association of these polymorphisms with rheumatic fever among Egyptian cases. In a case–control study, we attempted to test the association of susceptibility and severity of RHD with polymorphisms of two proinflammatory cytokines (TNF-α at position −308 and IL-6 at position −174) and two anti-inflammatory cytokines (IL-10 genes at position −1082 and IL-1Ra VNTR) among Egyptian affected cases.

Patients and Methods

This study included 50 children with RHD recruited from the pediatric cardiology department of Mansoura University Children Hospital, which is the main referral hospital in the Nile Delta region of Egypt. Their presenting diagnosis was based on revised Jones criteria by examination and investigation, including anti-streptolysin O, CRP, ESR, ECG, and echocardiography. All patients had chronic RHD with residual valve affection at least after 2 years from the first attack. Their mean age was 12.2 ± 3.4 years (range, 5–18), and sex was 29 males and 21 females. Of these patients, 16 (32%) had a positive family history of rheumatic fever and 14 (28%) had a positive history of parental consanguinity. Patients were classified as having either mitral valve damage (MVD) (29 cases, 52%) or multivalvular lesions (MVL) (21 cases, 42%). Furthermore, severity of valve lesions was graded according to echocardiographic findings as mild (21 cases, 42%), moderate (15 cases, 30%), or severe (14 cases, 28%). Case genotypes were compared to those of 98 healthy unrelated adult volunteers from the same locality. Their mean age 44.9 ± 6.7 years, and there were 52 males and 46 females. The genotypes were taken from normal adults with negative family history of the disease to avoid selection bias of controls.

DNA Extraction and Purification

After obtaining informed consent from all cases and controls, venous blood samples (3 ml) were collected in tubes containing ethylenediamine tetraacetate (EDTA), and then DNA was extracted promptly using a DNA extraction and purification kit (Gentra Systems) according to the manufacturer’s instructions and stored at −20°C until use.

Polymerase Chain Reaction Amplification

Three single nucleotide polymorphisms (SNPs) were analyzed, including promoter sites TNF-α−308 G/A, IL−10−1082 G/A, and IL-6−174 G/C as well as IL-1RaVNTR, as previously described [4, 5, 33, 39]. For identification of SNPs related to TNF-α, IL-6, and IL-10 genes, polymerase chain reaction with sequence-specific primers (PCR-SSP) was used in two reactions employing a common forward and two reverse primers. For IL-1Ra VNTR polymorphism, a single PCR reaction was used employing a forward and a reverse primer. All primers, Taq polymerase, dNTP, and MgCl2 were purchased from QiaGene. The assay was performed in a Techne-Genius thermal cycler. Briefly, 100–500 ng of genomic DNA was added to 25 µl of reaction mixture containing 1 µM of each common/specific primer, 200 µM of each dNTP, and 1 U of Taq DNA polymerase (Table 1). We used master mixes for each primer type and also control DNA samples for confirmation of negative amplifications to obtain accurate subject genotyping.

Table 1 Primer sequences and PCR conditions of the studied cytokines genes
Full size table

Detection of Amplified Products

The entire reaction volume plus 5 µl of bromophenol blue track dye were loaded into 2% agarose gel (Boehringer–Mannheim) containing ethidium bromide. Gels were electrophoresed for 20 minutes at 200 V, photographed under ultraviolet light (320 nm), and then scored for the presence or absence of an allele-specific band. Figure 1 shows the amplified PCR products of TNF-α–308 G/A, IL-10−1082 G/A, and IL-6−174 G/C compared to size marker, whereas Figure 2 shows amplified alleles of the IL-1Ra VNTR region in intron 2 of the gene.

Fig. 1
figure 1

PCR amplification bands using SSP for TNF-α−308 (band size, 863 bp) showing positive bands for the G allele above and the A allele below in lanes 1, 2, and 4 giving G/A genotype, only positive band for the G allele above in lanes 6 and 7 giving G/G genotype, and only positive band for the A allele below in lane 5 giving A/A genotype; IL-6−174 (band size, 234 bp) showing positive bands for the G allele above and the C allele below in lanes 1, 4, 6, and 7 giving G/C genotype, and only positive band for the G allele above in lane 2 giving G/G genotype; IL-10−1082 (band size, 179 bp) showing positive bands for the G allele above and the A allele below in lanes 1, 4, 5, and 7 giving G/A genotype, and only positive band for the A allele below in lane 2 and 6 giving A/A genotype. Lane M, DNA size marker; Lane 3, negative control (with no DNA)

Full size image
Fig. 2
figure 2

PCR amplification bands for IL-1RaVNTR showing positive A1 (band size, 410 bp) and A2 (band size, 242 bp) in lanes 1, 4, and 7 giving A1/A2 genotype, and only positive A1 in lanes 2, 5, and 6 giving A1/A1 genotype. Lane M shows DNA size marker and lane 3 shows negative control (no DNA)

Full size image

Statistical Analysis

Data were processed and analyzed using the Statistical Package of Social Science (SPSS, version 10.0). The frequency of studied allelic polymorphisms among cases was compared to that of controls describing the number and percentage of each and tested for positive association using Fisher’s exact test (modified chi-square test) and odds ratio (OR) with a minimum level of significance of p < 0.05.

Results

Tables 2–5 indicate that total cases showed a significantly higher frequency of homozygous genotypes of TNF-α−308 A/A (OR = 5.7, p < 0.001), IL−10−1082 A/A (OR = 3.1, p < 0.05), IL-10−1082 G/G (OR = 5.2, p < 0.05), and IL-1Ra A1/A1 (OR = 2.2, p < 0.05). These genotypes possibly contribute to disease susceptibility.

Table 2 Frequency of IL-10−108 2 G/A genotypes and allelic polymorphisms among cases of rheumatic carditis compared to controls
Full size table
Table 3 Frequency of IL-6−174 G/C genotypes and allelic polymorphisms among cases of rheumatic carditis compared to controls
Full size table
Table 4 Frequency of TNF-α−308 G/A genotypes and allelic polymorphisms among cases of rheumatic carditis compared to controls
Full size table
Table 5 Frequency of IL-1Ra VNTR genotypes and allelic polymorphisms among cases of rheumatic carditis compared to controls
Full size table

Cases with MVD showed significantly higher frequency of homozygous genotypes TNF-α–308 A/A (OR = 3.7, p < 0.05), TNF-α–308 G/G (OR = 4.4, p < 0.05), IL-10−1082 G/G (OR = 7.8, p < 0.05), and IL-1Ra A1/A1 (OR = 3.4, p < 0.05). Cases with MVL showed significantly higher frequency of homozygous genotypes of TNF-α−308 A/A (OR = 10.6, p < 0.001) and IL-10−1082 A/A (OR = 5.2, p < 0.05). This genotype shows a possible susceptibility for multivalvular affection among cases of RHD.

Cases with moderate severity showed a significantly higher frequency of homozygous genotypes TNF-α−308 A/A (OR = 7.7, p < 0.05), TNF-α−308 G/G (OR = 5.3, p < 0.05), IL-10−1082 A/A (OR = 4.5, p < 0.05), and IL-10−1082 G/G (OR = 5.1, p < 0.05). Cases with severe lesions showed significantly higher frequency of only the homozygous A/A genotype of TNF-α−308 A/A (OR = 5.9, p < 0.05). This genotype shows a possible susceptibility for that form of disease severity.

On the other hand, total cases as well as all subgroups showed a relatively significantly lower frequency of heterozygous genotypes, including those of TNF-α−308 G/A, IL-10−1082 G/A, and IL-1Ra A1/A2. These genotypes could be considered low-risk or protective genotypes.

Regarding allele frequency, it was found that cases with MVD had significantly higher frequency of TNF-α−308 A allele (OR = 2.9, p < 0.05) and IL-1Ra A1 allele (OR = 11.75, p < 0.001), with significantly lower frequency of TNF-α−308 G allele (OR = 0.34, p < 0.05) and also the IL-1Ra A2 allele (OR = 0.1, p < 0.001) compared to controls.

The previous results were confirmed by studying composite genotypes of cases compared to controls and finding a significantly higher frequency of composite genotype of TNF-α−308 A/A with IL-10−1082 A/A (OR = 37.4, p < 0.001) followed by TNF-α−308 A/A with IL-10−1082 G/G (OR = 31.6, p < 0.001), TNF-α−308 A/A with IL-1Ra A1/A1 (OR = 7.23, p < 0.001), and IL-10−1082 A/A with IL-1Ra A1/A1 (OR = 7.2, p < 0.05). Composite heterozygous genotypes of TNF-α−308 G/A and IL-10−1082 G/A and IL-1Ra 1/2 showed a significantly lower frequency among cases as represented in Table 6.

Table 6 Composite cytokine genotypes among RHD cases compared to controls
Full size table

Interestingly, IL-6−174 G/C genotype as well as its allele polymorphisms showed no statistically significant difference between total cases or subgroups and controls.

Discussion

Many authors have reported the presence of inherited immunoregulatory dysfunction in individuals susceptible to developing rheumatic fever, including the possibility of stimulation of certain cytokines resulting in a specific clinical behavior [12, 13].

Rheumatic fever among Egyptians is inherited on a multifactorial basis, with a heritability of 30% and a high degree (60%) of positive parental consanguinity [34]. In our study, analysis of case history also showed a positive family history of the disease and parental consanguinity in approximately one-third of cases, indicating a possible genetic contribution to the disease. Consanguinity can contribute to the presence of more allelic homozygosity and the appearance of disease phenotype, especially in recessive traits. However, testing genotypic and allelic frequencies among consanguineous versus nonconsanguineous cases showed that there was no significant difference, indicating that susceptible genetic makeup is not restricted to consanguineous cases (data not shown).

TNF-α is apparently one of the cytokines with an active and prominent role in the pathogenesis of the rheumatic process. Blood mononuclear cell cultures from rheumatic children produced more TNF-α than those from controls. TNF-α level was found to be increased in the serum due to infiltration of the heart by these inflammatory cells. Local production of these cytokines promoted the induction of postinfection autoimmune myocarditis [10, 23, 26].

In a previous study on rheumatic fever in Egypt, cytokine mRNA serum levels of IL-1α, IL-1B, and TNF-α before and after treatment were increased with inherited tendency for overproduction of IL-1α through a dominant genetic control, which may be the case with IL-2 and possibly TNF-α [25]. However, significant changes were found in serum values of IL-6 and TNF-α in the acute phase, on day 7 of treatment, and after treatment of rheumatic fever [26].

An interactive role of both TNF-α and IL-10 has been investigated and documented in many other immune diseases [18, 19, 29]. In the current study, we found an interactive pattern of TNF-α−308 and IL-10−1082 genetic polymorphisms because total cases showed a significantly higher frequency of homozygous genotypes A/A and/or G/G of both of them and were considered high-risk genotypes for susceptibility to RHD. This was also applicable for subgroups, including cases with MVD and cases with moderate valve severity. On the other hand, cases with MVL and severe lesions showed only significantly higher frequency of A/A but not G/G homozygosity genotypes of both genes. Also, significantly higher frequency of TNF-α−308 A allele was found among cases with MVL, with significantly lower frequency of G allele. Therefore, A/A homozygosity for both TNF-α−308 and IL-10−1082 may potentially be related to susceptibility to a more severe form of the disease and multivalvular affection as well.

Our results were partially in accordance with those reported for RHD Mexican Mestizo patients with increased frequencies of TNF-α−308 A allele and decreased G allele, especially for cases with MVL with increased G/A and decreased G/G gneotypes [15]. Racial or ethnic differences may be an important explanation for variations of population polymorphisms. Again, determining the interactive polymorphisms of both TNF-α and IL-10 genes is important to obtain the complete picture of such situation. On the other hand, the TNF-α−308 A+ allele was also reported to be a poor prognostic marker among cases with different inflammatory and immune disorders such as septic shock, meningococcal meningitis, and cerebral malaria [6, 25, 28].

Polymorphism related to the IL-1Ra gene VNTR in intron 2 was reported to be associated with many immune and inflammatory disorders among different ethnic populations, such as psoriasis, multiple sclerosis, systemic lupus erythematosus, and Sjögren’s syndrome; however, it was also reported not to influence the susceptibility to or severity of rheumatoid arthritis and Crohn’s disease[3, 37]. Among our Egyptian cases and controls, we found an IL-1Ra (VNTR) A1/A2 allelic ratio (79.4/20.6) similar to that reported internationally (73.6/21.4) [37]. RHD cases showed significantly higher frequency of the homozygous genotype A1/A1 with significantly lower frequency of the heterozygous genotype A1/A2 compared to controls, especially noted in MVD cases. On the other hand, it showed no relation to severity of rheumatic valve disease.

Polymorphism related to the −174 G/C in the promoter region of the IL-6 gene was reported to be associated with a variety of major diseases, such as Alzheimer’s disease, cancer, non-insulin-dependent diabetes mellitus, sepsis, and systemic-onset juvenile chronic arthritis. However, authors of previous in vitro and in vivo studies have reported conflicting results regarding the functionality of this polymorphism [2, 16, 30, 31]. In our study, we found no significant difference in the frequency of IL-6−174 polymorphic genotypes between cases and controls. In a previous study in Egypt, no increase in the production of IL-6 and IL-10 in acute rheumatic carditis was found [12].

The interactive role of cytokine polymorphisms may be through affection of the proinflammatory/anti-inflammatory balance presented within the Th1/Th2 balance. The perturbation of this balance in either direction could have major implications for the clinical course of many immune and infectious diseases [20].

We conclude that predisposition to RHD is influenced by genetic factors including cytokine gene polymorphisms with a possible susceptibility to a severe disease with multivalvular affection among cases with composite polymorphism TNF-α−308 A/A with IL-10−1082 A/A, followed by TNF-α−308 A/A with IL-10−1082 G/G. We recommend undertaking more family studies with linkage analysis to evaluate the actual risk among families with multiplex cases.