Introduction

Systemic lupus erythematosus (SLE) is a chronic, autoimmune disease which can damage many organs [1]. Despite of the survival of SLE patients has improved over the last 50 years, owing to the innovative therapies, the most appropriate scheme is absent. However, the first therapeutic option is still the glucocorticoids; treat-to-target therapy could improve the disease outcome supported by clinical trials. Lupus activity can be reduced by glucocorticoids, dramatically otherwise the side effects of long-term used glucocorticoids worsening quality of life [2]. Moreover, specific biomarkers that could predict the efficacy and side effects of glucocorticoids are still absent [3].

Both endogenous and exogenous glucocorticoids contribute to the down-regulation of disease activity and the outcome of organ damages in SLE via the intracellular glucocorticoid receptor (GR) [4]. Highly studied polymorphisms of the GR gene are associated with altered sensitivity to glucocorticoids [5]. The N363S and BclI polymorphisms have been associated with increased, whilst the ER22/23EK and A3669G polymorphisms have been associated with decreased glucocorticoid sensitivity [6].

The glucocorticoid receptor sensitivity can show a high degree of individual differences [7]. Furthermore, SNPs may alter the sensitivity for endogenous glucocorticoids, therefore may have significance in the pathomechanism of SLE [8]. Our aim was to assess the association between the BclI, N363S, and A3669G GR gene polymorphisms, and parameters and clinical manifestations of SLE.

Materials and methods

Patients

GR gene polymorphisms were analysed in 104 patients diagnosed and regularly followed-up tracked with SLE at the National Institute of Rheumatology and Physiotherapy. The control group contained 160 healthy individuals from the Hungarian population. In the SLE group, patients who have been presented with glucocorticoid therapy were selected. In the SLE group the female to male ratio was 89%, whilst in the control group was 69.37%. The patient’s average age was 47.9 ± 13.1 years, whilst in the control population this ratio was 52.73 ± 14.7 years. The average age at the diagnosis of SLE was 31.1 ± 13.2 years (Table 1).

Table 1 Demographic characteristics of the study population

All members of the SLE and healthy control groups were Caucasians origin. The research was approved by the Local Ethical Committee of Semmelweis University (SE TUKEB 12/2013). Agreed written consent was obtained from all patients.

Clinical and immunoserological analyses

SLE was classified as per the more recent 2012 SLICC-ACR (Systemic Lupus Collaborating Clinics revised and validated the American College of Rheumatology) criteria [9]. Particular manifestations of SLE—if have been ever presented from the onset of the disease—were selected and defined as in the classification criteria. Regarding neuropsychiatric manifestation we used the ACR Ad Hoc Committee on neuropsychiatric lupus nomenclature as providing a definition of 19 manifestations [10].

In addition the following immune serology parameters were tested in all patients: anti-dsDNA antibody, anti-ribosomal-P-protein antibody, anti-chromatin antibody, anti-C1q antibody, anti-SSA, and anti-SSB were tested by ELISA (ORGENTEC Diagnostika GmbH, Mainz, Germany). Anti-Sm antibody, anti-cardiolipin antibody IgG, anti-beta2-GPI antibody IgM, and IgG were tested by ELISA (INOVA Diagnostics, San Diego, CA, USA). Serum complement C3 and C4 were tested by nephelometry (Siemens Healthcare Diagnostic Products GmbH, Marburg, Germany). Lupus anti-coagulant was tested according to international recommendation.

DNA extraction and genotyping of GR gene polymorphisms

Genotyping of the BclI, N363S, and A3669G polymorphisms was performed in peripheral blood DNA isolated with commercially available DNA Isolation Kit (QIAamp DNA Blood Mini Kit (QiampDNA Blood Kit, Qiagen, USA). Genotypes for the BclI and the N363S variants were determined by allele-specific polymerase chain reaction (PCR) as earlier reported [11,12,13]. Genotypes for the A3669G polymorphism were analysed using a primer-probe set purchased as predesigned Taqman allelic discrimination assay according to the manufacturer’s instructions (Applied Biosystems, Applied Biosystems Group 850 Lincoln Center Drive Foster City, CA) on a 7500 Fast Real-Time PCR System (Applied Biosystems). Genotypes of the GR gene, BclI, N363S, and A366G, were compared between SLE patients to a control group consisting of 160 healthy individuals.

Statistical analysis

For statistical analysis Statistica software (7.0 version, Statsoft Inc.) was used. The Hardy-Weinberg balance was tested and did not show alteration for any polymorphisms. The differences between allele frequencies and prevalence of various symptoms were evaluated by chi-squared or Fischer exact tests. The demographic data were analysed by Student’s t test. Significant results were defined if p value was less than 0.05.

Results

Demographic findings and immunoserological parameters

The immunserological parameters of patients showed high degrees of diversity. The anti-Sm (referring for kidney and neuropsychiatric manifestation) and anti-SSA antibody (showing subacute cutaneous erythematosus) positivities were 25.96% and 39.42%, respectively. The frequency of anti-cardiolipin and anti-β-2-glikoprotein I antibodies and lupus anti-coagulant were as follows: 31.73%, 22.12%, and 21.15%. The results of the main demographic findings are summarised in Tables 1 and 2.

Table 2 Immune serologic findings in SLE patients

The allele frequency of BclI, N363S, and A3669G of the GR gene polymorphisms in patients with systemic lupus erythematosus and the control population

The occurrence of BclI polymorphism in patients with SLE was significant lower than control population (0.26 vs. 0.35, p = 0.025). The frequency of N363S and A3669G polymorphisms did not alter in the patients and healthy controls (N363S 0.03 vs. 0.03, p = 0.873; A3669G 0.16 vs. 0.22, p = 0.179) (Table 3).

Table 3 Allele frequency in SLE patients and control population

The association between BclI, N363S, and A3669G GR gene polymorphisms and clinical parameters of patients with systemic lupus erythematosus

Associations between of BclI polymorphism and the clinical symptoms

There was a significant association between of the psychiatrics symptoms and the carrier status of BclI polymorphism (p = 0.02), whilst a tendency (p = 0.06) with central nervous system symptoms. Patients with BclI polymorphisms suffered from neuropshichiatric symptoms more often than patients without BclI polymorphism (Table 4). No statistically significant differences were found between the BclI polymorphism and other clinical parameters of SLE.

Table 4 Association of the BclI polimorphism and neuropsychiatric symptoms

Associations between the N363S polymorphism and clinical symptoms

No statistically significant association were detected between the N363S polymorphism carrier status and SLE symptoms.

Association between the A3669G polymorphism and the clinical symptoms

Impact of A3669G polymorphism contributed a strong association with the psychiatric symptoms. Contrary to the association between BclI polymorphisms and psychiatric symptoms, these symptoms occurred less frequent in patients who carried the A3669G SNP compared to noncarriers (Table 5). No other significant association between this SNP and clinical parameters were detected (Table 6).

Table 5 Association of A3669G polimorphism and psychiatric symptoms
Table 6 Statistical analysis of GR gene polymorphisms (BclI, N363S, A3669G) and clinical parameters of systemic lupus erythematosus

Discussion

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease with several clinical manifestations and immunological alteration [1]. The interaction of sex, environmental factors, and hormonal status, including especially the hypothalamo–pituitary–adrenal axis (HPA) alteration, contribute to the development and existing of the disease. SLE is characterised by the loss of immune tolerance, aberrations of communication, differentiation, and activation of immune cells and pathways [1, 2]. Besides regulating metabolism, blood pressure, inflammation, and many other processes, changes in the cortisol level influence the cognitive function, the behaviour, and the reaction to stress [14]. We hypothetised that polymorphisms of glucocorticoid receptor gene, determining augmented or suppressed sensitivity to glucocorticoids, may have pathogenic significance in the development of SLE and may alter the clinical pattern of the disease.

We have found that the prevalence of BclI polymorphism was lower in the SLE group as compared to healthy controls. These findings are in concordance with the fact that BclI polymorphism increases glucocorticoid sensitivity and may have a preventive role in the development of SLE. One interesting association was found, however, between carrier status of BclI and neuropsychiatric symptoms. These symptoms developed more frequently in the BclI-positive SLE group, and in addition, these symptoms were less prevalent in the A3669G carriers. These findings strongly agree with an earlier study of van Oosten and colleagues [15]. The BclI polymorphism increases whilst the A3669G polymorphism decreases the sensitivity of glucocorticoid receptors and may have influence on the development of psychiatric syndromes.

In SLE, genetic susceptibility and other factors also trigger one of the most frequent symptoms, particularly the depression [16]. Decreased level of the noradrenalin, serotonin, and dopamine could stand at the background of depression by biogenic amine hypothesis [17]. Also, the dysregulation of the steroid level, therefore the stress, could be a prominent factor in the development of depression and cognitive dysfunction [18]. Normally, there are some molecular patterns characteristically recognised under stress situations. These patterns trigger the glucocorticoid secretion by the HPA axis and reduce the inflammatory process by negative feedback [19]. Symptoms of patients with depression show a positive correlation with the daily circle of cortisol. In the early morning cortisol triggers the symptoms but there is a behavioural relief in the evening hours [20]. In major depression changing GR sensitivity of the immune cells could lead to the dysregulation of the HPA-axis negative feedback. The alteration in the sensitivity of the receptors seems to be a critical point in the pathomechanism, whilst the role of the changed factors of the decreased expression of the receptors is controversial [21]. The altered tissue-specific cortisol level has a prominent influence on the mood, cognitive functions, the behaviour, and the reaction to stress [22]. Patients with bipolar affective disorders react intensively to the stress even in remission. Therefore, the insufficient anti-inflammatory response results increased concentration of inflammatory parameters. Moreover, there is a strong association between the increased number of the inflammatory cytokines produced by exaggerated microglial cell activity and the psychiatric disorders [23]. These cytokines could influence nerve cells producing serotonin and glutamate and resulting in a dysregulation of the homeostasis and lead to the depression. Taken together, there is a strong importance of the HPA-axis regulation and the GR sensitivity for cortisol behind the neuroimmunological mechanism [19, 20].

The elevated cortisol concentration could play also an important role in some other neuropsychiatric diseases [24]. There are a strong association between elevated cortisol level and the enlarged adrenal gland, therefore the HPA-axis hyperactivity, and psychosis. Also, there is a correlation between the elevated cortisol and the changes of the hypocampal structures in patients suffered by psychosis [23, 24]. Furthermore, there are coexistence between the concentration of the cortisol and the epileptic seizure onset [25,26,27]. An elevated corticosteroid level has been measured in meningoencephalitis compared to healthy population [28]. In Guillan-Barré syndrome, the changes of cortisol could influence the outcome of the disease [29].

In our research, we have found an association of the BclI polymorphism with the SLE patients suffered by psychiatric symptoms. We share Vivlar et al. agreement about that it is an important question to consider if the nature of the NP manifestations is vascular or inflammatory. The neuropsychiatric nomenclature of SLE (NPSLE) diagnosis may be added by presence of disease activity, presence of autoantibodies, and other confounding factors [30,31,32]. We have selected the neurological and other manifestations under analysis if have been ever presented. The average disease duration was 20.12 (2.6 ± 14.3) years, and in this period, the clinical symptoms usually have already appeared—especially in the first 5 years—in SLE. Therefore, new neurological symptoms are unexpected in further follow-up.

In contrast, there was a negative correlation between of the A3669G and SLE patients with psychiatric symptoms. Spijker et al. have found that A3669G SNP was also attributed a protective role in bipolar depression [33]. Unfortunately, we could not analysis the coherence of the manifestations (targeting mood disorders, stroke, psychosis, cognitive impairment, seizures, or headaches) and GR SNPs as the low eligible numbers of patients in the different classified subgroups of patients with CNS manifestation. Although, there were no difference between the SLE and control population in the prevalence of N363S and A3669G alleles.

To summarise, we found that BclI and A3669G polymorphism could influence the development or modulate the neuropsychiatric disorders in patients with SLE. Recent results suggest that BclI, and A3669G polymorphisms are more frequent in patients with neuropsychiatric manifestations refers to higher sensitivity to endogenous GC regulated by the tissue-specific gene expression. Moreover, results of this study could have some therapeutical consequences as exogenous GC could be effective in controlling SLE; however, it can be used only with caution when psychiatric manifestations are present.