Investigation of CTLA-4 and CD86 gene polymorphisms in Iranian patients with brucellosis infection



The protective immune response against Brucella involves T-cell-mediated immunity. T-lymphocyte receptors, CD28 and cytotoxic T-lymphocyte-associated protein-4 (CTLA-4), bind the same ligands, CD80 (B7-1) and CD86 (B7-2) on antigen-presenting cells and regulate T cell activation. CD28 delivers stimulatory signals whereas CTLA-4 provides inhibitory signals for T cell activation. Here, we investigated the association of four polymorphisms in CTLA4 (+49A/G [rs231775] and −318 C/T [rs5742909]) and its ligand CD86 (+1057 G/A [rs1129055] and +2379G/C [rs17281995]) with brucellosis infection. The study included 153 Iranian patients with active brucellosis and 128 healthy individuals as the control group. Genotyping of the CTLA4 and CD86 variants was performed using tetra amplification refractory mutation system-polymerase chain reaction (T-ARMS-PCR) and PCR–restriction fragment length polymorphism analysis, respectively. It was found that the CTLA4 −318 CT genotype and T allele were present more frequently in cases than in controls and are therefore associated with an increased risk for brucellosis (−318 TT genotype; OR = 2.544, P = 0.002). Likewise, the CD86 +1057 GA and AA genotypes and A allele were associated with an increased risk of brucellosis (+1057 AA genotype; OR = 3.81, P = 0.001). However, no statistically significant difference between brucellosis patients and controls in the allele and genotype distributions of CTLA4, +49A/G (P = 0.859) and CD86, +2379G/C (P = 0.476) was found. In conclusion, CTLA4 −318 CT genotype and T allele and the CD86 +057 GA and AA genotypes and A allele play roles as risk factors for developing brucellosis infection in Iran.

List of Abbreviations

antigen-presenting cell


allele-specific polymerase chain reaction


chronic obstructive pulmonary disease


cytotoxic T-lymphocyte-associated protein 4




linkage disequilibrium


odds ratio


polymerase chain reaction–restriction fragment length polymorphism


rheumatoid arthritis


single nucleotide polymorphism


tetra amplification refractory mutation system-polymerase chain reaction

Brucella spp., short, non-motile, non-sporulating, non-encapsulated, gram-negative aerobic rods, are crucial facultative intracellular pathogens of humans and livestock [1]. Brucellosis is the most common bacterial zoonotic disease worldwide, over half a million people being infected annually [2]. Although brucellosis has been controlled in many countries, it remains endemic in the Mediterranean and Middle Eastern regions including Iran, Turkey and the Arabian Peninsula [3-7]. Brucella spp. cause brucellosis with the pathophysiological manifestations of arthritis, endocarditis and meningitis in humans, and spontaneous abortion in cattle [8]. Brucella organisms invade cells of the reticuloendothelial system, and can be sequestered in macrophages at specific locations within the body, such as spleen, brain, joints, heart, liver, and bone marrow [9].

Both cell-mediated immunity and humoral responses are responsible for eradicating Brucella infection. Host protection against Brucella spp. depends primarily on cell-mediated immunity, involving mainly activated APCs such as macrophages and dendritic cells and CD4+, CD8+ T-lymphocytes [6]. Brucella antigens induce production of Th1 cytokines in humans [10, 11], and the Th1 immune response has been proved to be essential for eradication of Brucella infection [8]. Indeed, the interaction of co-stimulatory and co-inhibitory receptors (e.g., CD28 and CTLA-4) expressed on T-cells with the ligands (e.g., CD80 and CD86) on APCs influences the magnitude and duration of antigen-specific T-cell responses [12, 13].

CD86 (B7.2) is one of the essential co-stimulatory molecules expressed on APCs; others include receptor proteins, cytokines, and associated factors [14]. CD86 activates T-cells and regulates impaired dendritic cells [15], B-lymphocytes [16], and macrophages [17]. It has been demonstrated that Brucella abortus causes upregulation of CD86 in monocytes from healthy individuals [18, 19]. Activated T-cells express CD86 receptor-CTLA-4, which is both a homolog for CD28 and a member of the immunoglobulin superfamily. CTLA-4 (CD152) binds to both CD80 and CD86 with 20- to 100-fold higher affinity than CD28 and transmits an inhibitory signal, which opposes the action of CD28 on T-lymphocytes. CTLA-4 and CD28 bind the same ligands (CD80 and CD86); however, CD28 delivers positive whereas CTLA-4 provides inhibitory co-stimulatory signals [20]. Because CTLA-4 preferentially binds to B7 molecules, its inhibitory interaction eventually predominates, leading to downregulation of T-cell responses and immunological anergy as well as susceptibility to T-cell-mediated infectious diseases [17, 21]. Mice deficient in CTLA-4 have been shown to develop a severe lymphoproliferative disorder, autoimmune disease, and die [22]. Loss of CTLA-4 also leads to massive lymphoproliferation and fatal multi-organ tissue destruction, revealing a critical negative regulatory role for CTLA-4 [23].

The human CTLA4 gene maps to chromosome 2q33 and consists of four exons [24]. Polymorphisms have been identified in the CTLA4 and CD86 genes and have been associated with different susceptibilities to a wide range of T-cell-mediated autoimmune [25, 26] and infectious diseases [27, 28]. CTLA4 rs231775 (+49A/G) polymorphism leads to an alanine to threonine amino acid substitution of codon 17 of the leader peptide (A17T) [29] and CTLA4 rs5742909 is a C to T transition at position −318 (C-318T) of the promoter sequence [30]. Several reports strongly suggest that molecular variants in CTLA-4, +49A/G and −318 C/T are involved in T-cell-mediated autoimmune and infectious diseases such as type 1 diabetes [31], autoimmune thyroid disease [32, 33], multiple sclerosis [34], systemic lupus erythematosus [35], rheumatoid arthritis (RA) [26], primary biliary cirrhosis [30], hepatitis B virus [36], hepatitis C virus [37], Paracoccidioides brasiliensis [38] and Mycobacterium tuberculosis infections [39-41].

The CD86 gene has been mapped to 3q21; this locus consists of receptor proteins, cytokines and associated factors [19]. +1057G/A polymorphism in exon 8 (rs1129055) results in an alanine to threonine substitution at codon 304, which is located in the cytoplasmic tail of CD86 and contains putative phosphorylation sites for protein kinase C. This substitution has been shown to modify the degree of phosphorylation in this region and influence the APC-signal transduction pathway [42]. Another CD86 SNP, +2379C/G polymorphism (rs1721995) in the 3′-UTR of CD86, is reportedly located in the seed regions of miRNA binding sites and has a functional role in regulating expression of CD86 [26].

Functional CTLA4 and CD86 variants may contribute to the pathogenesis of disorders characterized by abnormal T-cell responses, including brucellosis. However, these genes have not been previously examined with respect to brucellosis susceptibility. Therefore, we designed this study to test the hypothesis that four SNPs (rs5742909 and rs231775 in CTLA-4, and rs17281995 and rs1129055 in CD86) are associated with brucellosis.


Study subjects

For this case–control retrospective study, 153 patients (102 men and 51 women) with active brucellosis (age range 6–76 years; mean ± SD, 31.24 ± 16.6) and 128 healthy individuals as a control group (93 men and 35 women, age range 19–64 years; mean ± SD, 34.04 ± 13.69) were recruited. Blood samples were taken from all participants and placed in EDTA-containing tubes for DNA extraction. All the patients were either milk farmers (had contact with diagnosed infected animals) or had a history of consuming raw milk and unpasteurized dairy products. Table 1 shows clinical characteristics and complications of brucellosis patients. Brucellosis was diagnosed on the basis of clinical manifestations (including fever, night sweats, weakness, malaise, weight loss, splenomegaly, lymphadenopathy, myalgia and arthralgia), serological tests, PCR assays and blood cultures. The control group was composed of healthy blood donors with no history of brucellosis or genetic disorders and was matched for age, sex, and geographic area. They had the same background as the cases and were at the same risk of exposure to brucellosis. Ethical approval for this study was obtained from the Ethics Committee of Shahid Sadoughi University of Medical Sciences and informed consent was obtained from all participants.

Table 1. Clinical characteristics of brucellosis patients
 Number (%)
Age, years31.24 ± 16.60
Men102/153 (66.66%)
Women51/153 (33.33%)
Fever99/153 (64.70%)
Myalgia38/153 (24.83%)
Anorexia85/153 (55.55%)
Headache58/153 (37.90%)
Malaise70/153 (45.75%)
Low back pain35/153 (22.87%)
Fatigue65/153 (42.48%)
Sweating93/153 (60.78%)
Weight loss53/153 (34.64%)
Arthralgia84/153 (54.90%)
Paresthesia29/153 (18.95%)
Palpitations26/153 (16.99%)
Nausea23/153 (15.03%)
Rash18/153 (11.76%)
Dysuria17/153 (11.11%)
Blood culture (positive)105/153 (68.62%)
Brucella species
Brucella melitensis114/153 (74.17%)
Brucella abortus38/153 (25.83%)1
Clinical complications
Arthritis22/153 (14.37)
Endocarditis2/153 (1.30)
Spondylitis4/153 (2.61)
Neurobrucellosis7/153 (4.57)
Meningitis1/153 (0.65)
Mortality6/153 (3.92)

Culture and identification of organism

Brucella strains were grown on 5% sheep blood-agar plates and incubated at 37°C in the presence of 5–10% CO2 for 48 hr. Macroscopically, typical and well-isolated Brucella-like colonies are small, transparent, raised, and convex, with complete edges and smooth and glistening surfaces along the streak lines. Brucella species, gram negative coccobacilli, were characterized microscopically using Gram staining. In addition, oxidase, catalase, urease, and other biochemical reactions were performed to identify Brucella species [43]. In patients whose blood cultures were negative, a combination of PCR assay of Brucella 16s rRNA gene (EMBL accession no. X13695), clinical manifestations, positive serological tests (defined as Wright titer ≥1/160 plus mercaptoethanol test ≥1/80 or Coombs Wright ≥1/320) provided adequate confirmation of brucellosis infection. Of the 153 brucellosis patients, 105 had positive blood cultures. The subjects with negative blood cultures were tested for brucellosis by PCR assay based on the protocol of Mukherjee et al. [44]. PCR assay of the 16s rRNA gene showed that all 48 clinical strains belonged to the genus Brucella.

Genotyping of CTLA4 (+49A/G, −318 C/T) and CD86 (+2379G/C, +1057 A/G) variants

Genomic DNA was extracted from the peripheral blood leukocytes by a “salting-out” method, as described previously [45]. The quality of the isolated DNA was checked by electrophoresis on 1% agarose gel, quantitated spectrophotometrically and stored at −20°C till further use. CTLA-4 +49A/G (rs231775) exon 1 polymorphism and the CTLA-4 −318 C/T (rs5742909) promoter polymorphism were genotyped by T-ARMS-PCR, as described previously [46], using four primers, two outer primers (FO and RO) and two inner allele specific primers (FI and RI). Cycling condition for CTLA4 +49 A/G were initial denaturation at 95°C for 5 min followed by 35 cycles of 30 s at 95°C, annealing temperature for 30 s at 62°C, and a final cycle at 72°C for 7 min [46]. The PCR conditions for CTLA4 −318 C/T polymorphism were 5 min at 95°C followed by 35 cycles of 95°C for 30 s, 58°C for 30 s, 72°C for 30 s followed by a final extension step for 10 min at 72°C [47]. PCR (total volume 20 mL) was performed using commercially available Prime Taq premix (Genetbio, Nonsan Si, South Korea) according to the manufacturer's instructions. PCR products were separated by standard electrophoresis on 2.5% agarose gel containing ethidium bromide. T-ARMS-PCR primers and products size are shown in Table S1.

CD86 +2379G/C (rs17281995) 3′-UTR polymorphism and CD 86 +1057G/A (rs1129055) exon 8 polymorphism were genotyped using PCR–RFLP analysis as described by Xiang et al. [47]. In brief, PCR was carried out using the following conditions: 95°C for 5 min, 35 cycles of 95°C for 30 s, 62°C (for rs17281995)/60°C (for rs1129055) for 30 s, 72°C for 30 s, followed by a 10-min extension at 72°C. The PCR products were further digested with restriction enzymes, BbvI or MlyI, in a 10 µL reaction volume at 37°C for 4 hr. RFLP–PCR primers, conditions, amplicons size and restriction enzymes are presented in Table 1.

Statistical analysis

All statistical analyses were performed using SPSS software for Windows, version 18.0 (SPSS, Chicago, IL, USA). The association between genotypes and brucellosis was assessed by computing the OR and 95% CI by logistic regression analyses. P-values below 0.05 were considered statistically significant. The Hardy–Weinberg equilibrium was tested with the χ2 test for any of the SNPs under consideration. LD and frequencies of haplotypes in the controls and patients were estimated using SNPStats software [48].


Genotype frequencies of CTLA-4 and CD86 polymorphisms

These four SNPs were successfully genotyped in 153 brucellosis patients and 128 control subjects. Genotype distributions of these four SNPs in our control subjects agreed with that expected under Hardy–Weinberg equilibrium. The genotype and allele frequencies of these four SNPs in the studied groups are presented in Table 2.

Table 2. The genotype and allele frequencies of CTLA4 and CD86 SNPs according to brucellosis patients and controls
PolymorphismAllele/genotypeBrucellosis patients n = 153 (%)Controls n = 128 (%)OR (95% CI)P-value
  1. Ref., reference.
+49A/G (exon 1)
AlleleA199 (65)163 (64.2)Ref.
G107 (35)91 (35.8)0.963 (0.680–1.363)0.859
GenotypeAA77 (50.3)53 (41.4)Ref.
AG45 (29.4)51 (39.8)0.607 (0.357–1.034)0.066
GG31 (20.3)24 (18.8)0.889 (0.470–1.682)0.718
−318 C/T (promoter)
AlleleC257 (84)234 (92.1)Ref.
T49 (16)20 (7.9)2.231 (1.287–3.864)0.004
GenotypeCC104 (68)108 (84.4)Ref.
CT49 (32)20 (15.6)2.544 (1.416–4.57)0.002
+2379G/C (3′ UTR)
AlleleG274 (89.5)234 (91.4)Ref.
C32 (10.5)22 (8.6)1.242 (0.702–2.196)0.476
GenotypeGG121 (79.1)106 (82.8)Ref.
GC32 (20.9)22 (17.2)1.274 (0.698–2.327)0.430
+1057G/A (exon 8)
AlleleG130 (42.5)141 (55.1)Ref.
A176 (57.5)115 (44.9)1.659 (1.877–2.319)0.003
GenotypeGG17 (11.1)34 (26.6)Ref.
GA96 (62.7)73 (57)2.63 (1.364–5.073)0.004
AA169 (26.2)21 (16.4)3.81 (1.736–8.361)0.001

The allele and genotype frequencies of both the CTLA4, +49A/G (P = 0.859) and CD86, +2379G/C (P = 0.476) variants did not differ significantly between the brucellosis patients and controls. However, a significant association with the genotypes and alleles of CTLA4 −318 C/T and CD86 +1057G/A variants was found in the brucellosis cases compared with the controls. The CTLA4 −318 CT genotype and T allele were risk factors for brucellosis, having a higher prevalence in cases than in controls (OR = 2.544, 95% CI = 1.416–4.57, P = 0.002 for the CT genotype; OR = 2.231, 95% CI = 1.287–3.864, P = 0.004 for the T allele). Likewise, the CD86 +1057 GA and AA genotypes and A allele were associated with increased risk of brucellosis, having a higher frequency in patients than in controls (OR = 2.63, 95% CI = 1.364–5.073, P = 0.004 for the GA genotype; OR = 3.81, 95% CI = 1.736–8.361, P = 0.001 for the AA genotype; OR =1.66, 95% CI = 1.877–2.319, P = 0.003 for the A allele).

Analysis of LD and haplotype association of CTLA-4 and CD86 polymorphisms

LD was tested by calculating Lewontin's D′ coefficient and the correlation coefficient r2 [49]. Pairwise LD between the SNPs in CTLA-4 (rs5742909, rs231775) and CD86 (rs17281995, rs1129055) was calculated for cases and controls. LD analysis of the CTLA-4 SNPs demonstrated incomplete LD for each pair of SNPs in controls and brucellosis patients (rs5742909/rs231775, D′ = 0.74, r2 = 0.141) and CD86 (rs17281995/rs1129055, D′ = 0.961, r2 = 0.091).

Table 3 shows haplotype association analyses for rs231775 and rs5742909 of the CTLA4 and rs17281995 and rs1129055 of the CD86 in brucellosis patients and controls. The rs231775/rs5742909 AT and rs17281995/rs1129055 GA haplotypes were found to be risk factors for brucellosis, occurring with greater frequency in patients than in controls (P = 0.002 and P = 0.007, respectively). However, the rs17281995/rs1129055 GG haplotype, which had a higher frequency in controls than in cases (P = 0.003), was identified as a protective factor against brucellosis.

Table 3. Results of haplotype association analyses for rs231775 and rs5742909 of CTLA4 and rs17281995 and rs1129055 of CD86 in brucellosis patients and controls
HaplotypeCase (frequency)Control (frequency)OR (95% CI)P-value
  • All frequencies <0.03 were ignored in this analysis.
CTLA4 SNPs (rs231775/rs5742909)
AC187 (0.612)163 (0.642)0.881 (0.624–1.243)0.469
AT11 (0.038)0 (0)0.002
GC693 (0.228)71 (0.280)0.761 ([0.519–1.115)0.161
GT37 (0.122)20 (0.079)1.622 (0.917–2.871)0.094
CD86 SNPs (rs17281995/rs1129055)
CA31 (0.101)22 (0.086)1.203 (0.678–2.136)0.526
GA145 (0.474)93 (0.363)1.588 (1.131–2.231)0.007
GG129 (0.422)141 (0.551)0.598 (0.428–0.836)0.003
CG1 (0.003)0 (0.000)


Brucellosis is an intracellular granulomatous bacterium with worldwide distribution. Adequate Th1 immune responses are fundamental for eradication of Brucella infection. Brucella antigens induce production of Th1 proinflammatory cytokines (IFN-g and TNF-α) in humans [50, 51]. Recently, it has been demonstrated that the CD28:B7 pathway is important in IFN-g and TNF-α production by peripheral blood mononuclear cells from brucellosis patients [52]. The role of this pathway in the immune response is essential, because a signal through CD28 stimulates IL-2 production [11] and IL-2R upregulation, leading to T cell proliferation and activation, a process essential to control Brucella multiplication [19]. CTLA4, a member of the immunoglobulin superfamily, is a key component in the immune system. CTLA-4 is expressed exclusively on activated CD4+ and CD8+ T cells and binds the same ligands, CD80 (B7-1) and CD86 (B7-2), as CD28 but with a 20- to 50-fold greater affinity [24, 53, 54]. CTLA-4 induces immune tolerance and is one of the critical negative regulators of the T-cell-mediated immune response [55]. Two mechanisms for inhibition of T-cell-mediated immune response by CTLA-4 have been suggested: One is delivery of a negative signal to T-cells, with rapid inhibition of T-cell activation, and the other is B7 (CD80, CD86) sequestration on APCs and induction of T-cell anergy [56]. Consequently, blockade of the CD28:B7 pathway by CTLA4 results in downregulation of proinflammatory cytokines, thus possibly increasing the risk of T-cell-mediated diseases [36]. Thus, any variations in CTLA4 expression or function perhaps contribute to an increased or decreased risk of brucellosis.

Our study provided evidence for a genetic association between CTLA −318 C/T variant and increased risk of brucellosis. We found that the T allele occurs with greater frequency in brucellosis patients than in controls, indicating it is a risk factor for this disease. This finding is in accordance with several studies, which have shown associations between this variation and risk of COPD [17], Graves disease [57], autoimmune type 1 diabetes [54], acute anterior uveitis [58] and multi-factorial autoimmune diseases [59]. However, this association was not found for Behcet disease [60], RA [61], and systemic lupus erythematosus [62]. The TT genotype at −318 C/T has been shown to be associated with upregulation of CTLA4 transcription [63]. It has been demonstrated that −318 TT genotype, which correlates with CTLA4 high producer genotype, is associated with reduced control of T-cell activation and in this way probably contributes to the pathogenesis of brucellosis.

Another SNP that showed significant association with Iranian brucellosis patients was +1057G/A, rs1129055 in the CD86 gene. In this study, we found that the A allele of +1057G/A may enhance the risk of brucellosis by 1.6-fold. This finding indicates that CD86 +1057G/A polymorphism may also confer susceptibility to brucellosis. Consistent with our finding, this polymorphism has been associated with increased risk of COPD [17] and pancreatic cancer [47]. Liu et al. [17] found that the A allele of rs1129055 augments the risk of COPD by 1.3-fold. Marin et al. [64] have also shown that rs1129055 polymorphism is involved in liver transplant acceptance, possibly decreasing acute rejection frequency and increasing graft survival. However, the rs1129055 variation has not been found to be associated with a number of diseases including RA [26, 65], coronary artery disease [42], and Ewing sarcoma [66]. The G-to-A transition at position +1057 has been shown to cause exchange of alanine against a threonine residue at codon 304, which introduces a potential phosphorylation site in the cytoplasmic region. This substitution may modify the degree of phosphorylation in this region and influence the APC-signal transduction pathway [17, 26, 66]. CD86, the major CD28 ligand, which is constitutively expressed on monocytes and activated B cells, activates T-cells and regulates the immune response [64]. Given that CD86 malfunction has been associated with impairments in dendritic cells, B-lymphocytes, macrophages and T-cell activation [64], the CD86 functional variation, +1057G/A, probably contributes to the pathogenesis of brucellosis in the Iranian population.

On the other hand, our study failed to show any association between CTLA4 + 49 A/G and CD86 +2379G/C variants and brucellosis risk. In agreement with our findings, these SNPs have not been associated with risk of COPD [17], RA [26], or systemic sclerosis [46]. However, these polymorphisms have been shown to be associated with primary biliary cirrhosis [67, 68] and hepatitis B infection [36].

In conclusion, our study is a preliminary report that provides evidence that CTLA4 −318 C/T and the CD86 +1057G/A polymorphisms act as risk factors for developing brucellosis infection in the Iranian population. However, our study is limited by a lack of the data regarding focal manifestations, relapses and responses to treatment. Replication of our findings in populations of different ethnic backgrounds is important to confirm the association between the disease and the specific polymorphism, regardless of the genetic background of the population.


The authors appreciate all individuals who willingly participated in this study. We thank R. Motamed-Zadeh for his kind help in sample collection.


The authors declare that they have no conflicts of interest to disclose.