Influence of prolactin and estrogen on disease activity in patients with systemic lupus erythematosus

Objective The objective of this paper is to evaluate the role of prolactin and estrogen levels on disease activity in patients with systemic lupus erythematosus (SLE). Patients and methods This study included 60 female patients with SLE, with a mean age of 33.5±13.12 years. It was conducted between November 2014 and October 2015. Disease activity was defined according to Systemic Lupus Erythematosus Activity Index; score of at least 6 was considered as an active disease. Prolactin (PRL) and estrogen levels and other serological markers of lupus disease activity, namely, complement 3,4 (C3 and C4), erythrocyte sedimentation rate, C-reactive protein, and anti-double-stranded DNA (anti-dsDNA) titer were calculated. Results Hyperprolactinemia was present in 25.0% of patients, and low estrogen level was present in 33.3% of patients. There was no significant correlation between either of estrogen or prolactin levels and all clinical and laboratory features, except for a significant positive correlation between anti-dsDNA and hyperprolactinemia. Conclusion There was no significant correlation between either of PRL or estrogen levels and Systemic Lupus Erythematosus Activity Index score. Overall, 80.0% of patients with hyperprolactinemia and 80.0% with low estrogen level had SLE activity. There was a significant difference in the frequency of further indicators of disease activity in SLE such as raised erythrocyte sedimentation rate, raised C-reactive protein, or decrease in complement factors with high serum PRL and low estrogen level.


Introduction
Systemic lupus erythematosus (SLE) is a potentially fatal and severe chronic autoimmune disease that affects multiple organ systems. It is remarkably heterogeneous, with diverse and dynamic symptoms manifested by flares of disease activity [1]. Hormonal, infectious, and environmental factors have been implicated in the etiology of the disease [2]. SLE is a disease of young women, which occurs from infancy to old age, with peak occurrence between ages 15 and 40 years. Females are affected far more than males (6-10 : 1) [3,4]. It seems highly plausible that female sex hormones contribute to the pathogenesis of lupus based on the tendency for disease onset during the child-bearing years [5], increased numbers of flares during high hormonal states such as pregnancy [6] and ovulationinduction therapy, and remissions after menopause [7]. It has been proved that steroid hormones such as 17 β-estradiol, testosterone, prolactin, progesterone, and dehydroepiandrosterone influence immune system regulation [8,9] and the activity of SLE [10].
Prolactin (PRL) participates in a number of important functions in the body: performs as a hormone, mainly owing to its pituitary production, and acts as a cytokine. Prolactin is also secreted by immune cells and its receptor belongs to the family of cytokine receptors type 1 [11], and it may play a role in the pathogenesis and clinical activity of SLE and other autoimmune diseases in human and experimental animal models [12]. PRL secretion is inhibited by the hypothalamus through dopamine. Thyroid-releasing hormone, hypothyroidism, and adrenal insufficiency stimulate PRL secretion by inhibiting dopamine secretion. The main cytokines stimulating PRL secretion are interleukin (IL)-1, IL-2, and IL-6, whereas interferon-γ and endothelin 3 are inhibitory [13].
The prevalence of hyperprolactinemia in the general population is lower than 5%. Nevertheless, the average prevalence of hyperprolactinemia in patients with lupus is 20-30%, varying from 8 to 69.7% [14]. There is controversy about the existence of a correlation between the disease activity and the concentration of serum PRL in SLE. Some authors reported these two parameters to be positively correlated [15,16], whereas others denied an association [17,18].
In the classical mechanism of steroid hormone action, estradiol diffuses into target cells and binds to estrogen receptors located in the nucleus. The ligand-activated receptors interact at specific DNA sites, termed estrogen response elements, along target genes and alter the rate of transcription [19]. Estradiol can both activate and repress genes within a given signal transduction pathway contributing to abnormal signal transduction in SLE T cells [20]. Grimaldi et al. [8] found that estrogens play an important role in B-cell maturation, selection, and activation and, thus, can potentially weaken the immune system. There are conflicting opinion in the literatures about the influence of estrogens on the development of SLE [8,9,[21][22][23].
We aimed in this study to evaluate the role of PRL and estrogen levels on disease activity in patients with SLE.

Patients and methods
This was a cross-sectional study, conducted between November 2014 and October 2015. It included 60 female patients with SLE aged from 16 to 58 years with a mean age of 33.5±13.12 years who were diagnosed according to the American College of Rheumatology Criteria [24]. The study was carried out with the approval of the responsible ethics committee and in accordance with national law and the Helsinki Declaration of 1975 (in its current, revised form). Informed consent was obtained from all patients.
Demographic characteristics, clinical manifestations, and autoantibody profile such as anti-double-stranded DNA (anti-dsDNA) were recorded. Disease activity was defined according to Systemic Lupus Erythematosus Activity Index (SLEDAI) [25]; score of at least 6 was considered as an active disease.
Besides clinical assessment, venous blood was taken for measurement of the PRL, estrogen levels, and other serological markers of lupus disease activity, namely, C3, C4, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and anti-dsDNA titer.
Patients with renal and/or hepatic failure, pregnancy, lactation, hypothyroidism, taking medications known to affect PRL level, or taking sex hormones (oral contraceptives, hormone replacement, etc.) were excluded from the analysis.

Samples collection
The patients under investigation were fasting. Overall, 10 ml of venous blood was taken from all patients between 8:00 and 10:00 AM and divided into the following: 3 ml in sterile EDTA-containing tubes for complete blood counts (CBC), 1.6 ml in tubes containing 3.8% sodium citrate for ESR, and the remainder was left in plain tubes for spontaneous clotting at room temperature before being centrifuged at 3000 rpm for 10 min. Serum samples were kept at −20°C for the hormonal determination (PRL and estrogen) and further analysis.

Laboratory tests
Kidney and liver function tests and lipid profile were performed using a chemical analyzer Hitachi 911 (Boehringer Mannheim, Germany). CBC was detected by Beckman Coulter (Brea, California, USA) HMX. ESR was performed using Westergren method.

Statistical analysis
Data were analyzed using the statistical package SPSS, version 21. Data were expressed as mean±SD, median, and frequencies. Group differences were compared by using t-test, one-way analysis of variance, ×2, Mann-Whitney U-test, and Fisher's exact test when applicable. Pearson's correlation coefficient (r) between variables was calculated.
Linear regression analysis was performed. Serum PRL level was used as the dependent variable. Probability levels below 0.05 were considered significant.
We had 20 (33.3%) patients with low estrogen level, 40 (66.7%) patients with normal estrogen level, but no patients with high estrogen level. Table 1 shows the demographic and clinical features of the normoprolactinemic and hyperprolactinemic groups of patients. Nonsignificant difference was observed between normal and hyperprolactinemic groups regarding different clinical features such as photosensitivity, malar rash, discoid rash, oral ulcers, arthralgia, arthritis, pluritis, nephritis, pericarditis, and central nervous system manifestations (P>0.05). Table 2 shows the laboratory and serological profiles of the normoprolactinemic and hyperprolactinemic groups of patients and SLEDI score. There was no significant correlation between PRL level and red blood cells, white blood cells, platelets, hemoglobin level, 24H protein, and creatinine clearance. However, a significant positive correlation was observed with   Table 3 shows the demographic and clinical features of the groups of patient with normal and low estrogen level. There was no significant correlation between the estrogen level and all clinical features (P>0.05). Table 4 shows the laboratory and serological profiles of the normal and low estrogen groups of patients and SLEDAI score. There was no significant correlation between estrogen level and red blood cells, white blood cells, platelets, hemoglobin level, 24H protein, creatinine clearance, and anti-dsDNA.
Correlation was also tested between the estrogen level and other serological markers of lupus activity. Again, no significant correlation could be demonstrated between estrogen level and C3, C4, and ESR. There was no significant correlation between estrogen level and SLEDAI score. Overall, 16/20 (80%) of low estrogen patients had SLE activity.
There was a significant difference in the frequency of several clinical manifestations and serological parameters between patients with SLE with normoprolactinemia and hyperprolactinemia (malar rash, discoid rash, nephritis, pluritis, pericarditis, and central nervous system manifestations). Also, there was a significant difference in the frequency of several clinical manifestation parameters between patients with SLE with normal and low estrogen level (malar rash, oral ulcers, arthralgia, arthritis, pericarditis, and elevation of  CRP). Low complement appeared to be more common in hyperprolactinemic and in low estrogen groups, although the difference did not reach statistical significance.

Discussion
SLE is a complex autoimmune disorder that develops in genetically prone individuals under the influence of various environmental factors. It is an autoimmune disease with a wide array of clinical manifestations. It is characterized by the production of antibodies to components of the cell nucleus [3,4]. We aimed in this study to evaluate the role of prolactin and estrogen levels on disease activity in patients with SLE.
Hyperprolactinemia in patients with SLE may be caused by either enhanced secretion of pituitary PRL under the effect of inflammatory cytokines [27] or increased production of PRL by peripheral lymphocytes [28]. The immune complexes of PRL-anti-PRL (which are the macroprolactins) are not biologically active, as their large size interferes with transversing the capillary walls to reach target tissues. Delayed clearance of the PRL-IgG complex may account for increased serum levels of PRL in these patients [29]. A high number of patients with SLE were found to be hyperprolactinemic [30], but there is a controversy about the existence of a correlation between disease activity and the concentration of serum PRL in patients with SLE.
Most studies point to a positive correlation between PRL and SLE activity [15, 16,31]. Yang et al. [32] proved that the serum level of PRL was increased in the patients with active SLE compared with patients with inactive SLE and that the serum level of PRL was closely related to SLE disease activity. In the study by Zahra et al. [33], mild to moderately elevated PRL levels were found in 10/30 (33.3%) of patients with SLE, and there was significant association between high PRL levels and clinical disease activity. Jacobi et al. [34] in their study revealed a positive correlation of the disease activity and the serum PRL concentration in patients with SLE. Patients with high disease activity had significantly higher serum PRL levels compared with patients with less active disease.
Other studies denied an association between PRL and SLE activity [35,36]. In a study by Pauzner et al. [37], mild hyperprolactinemia was found in 20/82 (24%) patients, and no association between hyperprolactinemia and clinical disease activity could be demonstrated. The study by Mansoor et al. [38] reveled that there was no significant correlation between serum PRL levels and SLEDAI and other serological disease markers, namely C3, C4, and ESR. These disagreeing results about the correlation between PRL and SLE activity can be explained by the heterogeneity of the groups of patients studied, by the use of different index to measure SLE activity, by the inclusion of patients with variable disease duration and by the diverse methodologies used for PRL testing. The presence of hyperprolactinemia is associated with diverse autoantibodies like antinuclear antibody, anti-dsDNA, anticardiolipin and antimicrosomal [39].
Antibodies directed toward nuclear antigens are characteristic of SLE, whereas anti-dsDNA antibodies are the hallmark for the disease [40]. Another finding suggesting a relationship between the serum PRL concentration and the disease activity in SLE is the positive correlation between the PRL level and the concentration of anti-dsDNA (IgG) [34]. Neidhart et al.
[15] have shown these autoantibodies to be positively correlated with the serum PRL concentration in patients with SLE. In addition, Miranda et al. [41] reported a trend of anti-dsDNA-positive patients with lupus nephritis to be hyperprolactinemic compared with those who did not have anti-dsDNA antibodies. Zahra et al. [33] demonstrated that hyperprolactinemia in a subset of their patients correlated with high serum level of anti-dsDNA. Yang et al. [32] showed that increased serum levels of PRL were related to immunoglobulin and anti-ds-DNA antibody production. After treatment, the serum level of PRL was decreased with the reduction in the anti-ds-DNA antibody titer. This suggest that serum PRL might affect B-cell activation and antibody production, and that PRL might be implicated as a modulator of humoral immunity. Zhu et al. [42] showed a positive correlation in serum PRL levels and specific antibodies against dsDNA.
In our study, a significant positive correlation was observed between the serum PRL concentration with anti-dsDNA (r=0.387, P<0.05).
In addition, in the study by Jacobi et al. [34], the frequency of further indicators of disease activity in SLE, such as raised ESR or decrease in complement factors, was associated with high serum PRL.
In the study by Zhu et al. [42], a negative correlation was found between serum PRL levels and complement component C3.
In our study, there was a significant difference in the frequency of further indicators of disease activity in SLE such as raised ESR, raised CRP, or decrease in complement factors between normoprolactinemic and hyperprolactinemic patients.The influence of estrogens on the development of SLE remains unclear. Some studies underline the negative influence of these hormones on the immune system [8,9,21,43], especially in patients with some genetic predisposition [6,7], whereas others show the positive influence on health [22,23]. In one study, the hormonal replacement therapy was associated with SLE development. No association was found when analyzing the risk for SLE among oral contraceptive users [44]. Others studies have reported that there was a group of female affected with SLE in which the use of hormonal replacement therapy or oral contraceptive did not cause SLE exacerbation, but in a few number of patients with SLE, it did [45,46]. There are different opinions about the disease activity influence on ovarian function. Some studies stressed the relationship between SLE activity and menstrual cycle disturbances; in other studies, this fact was not confirmed [47,48]. In patients with SLE, the aromatic hydroxylase activity was found to be increased, which may partially explain the abnormalities of peripheral estrogen metabolism observed in these patients [49]. Concerning serum 17 β-estradiol (E2), its levels were reported to be in increased, normal, or low in patients with SLE [43,50]. In the study by Shabanova et al. [51], the decrease of E2 level was dominant, and only in 2% of patients, its increase was observed. An investigation from Munoz et al. [50] obtained similar results: in SLE women during luteal phase of menstrual cycle, progesterone and E2 levels were decreased.
In our study, there were menstrual disturbances in 40/60 (66.6%) patients. We had 20 (33.3%) patients with low estrogen level, 40 (66.7%) patients with normal estrogen level, but no patients with high estrogen level.
There was no significant correlation between estrogen level and SLEDAI score; however, 16/20 (80%) of low estrogen patients and 30/40 (75%) of normal estrogen patients had SLE activity.
There was a significant difference in the frequency of further indicators of disease activity in SLE such as raised ESR, raised CRP, or decrease in complement factors between patients with normal and low estrogen level.
In conclusion, this study demonstrate that no significant correlation existed between either of hyperprolactinemia or estrogen level and SLE disease activity. However, 80% of hyperprolactinemic and 80% of low estrogen patients had SLE activity.
There was a significant difference in the frequency of further indicators of disease activity in SLE such as raised ESR, raised CRP, or decrease in complement factors with high serum PRL and low estrogen level, and a significant positive correlation was observed between anti-dsDNA and hyperprolactinemia.

Financial support and sponsorship
Nil.

Conflicts of interest
There are no conflicts of interest.