Hypothalamus-pituitary-adrenal axis (HPA)

The initial studies done at experimental level demonstrated that the susceptibility of Lewis rats (LEW/N) to an inflammatory/autoimmune disease similar to RA is due, at least in part, to an abnormal response in the HPA axis (CRH, ACTH, and glucocorticoids) (24).

Replacement of glucocorticoids in these rats reverses susceptibility to developing arthritis. HPA axis alteration in LEW/N rats is due to a decrease in the release of CRH which is genetically determined by the inhibition of its messenger RNA and alterations on pro-opiomelanocortin, globulin binding glucocorticoid, and its receptors (25). However, a normal response by HPA does not necessarily protect rats from the development of arthritis. As has been demonstrated in other experiments, rats exposed to high levels of stress develop arthritis despite high levels of cortisol (26).

In arthritis induced by immunization of rats with type II collagen, corticosterone levels rose only transiently, and adrenaline blood levels, hypothalamic IL-1beta and IL-6 overexpression rose during the induction phase of the disease. The increase in hypothalamic noradrenaline content during the symptomatic phase was paralleled by a gradual loss of sympathetic fibers in the joints. No correlation between hypothalamic IL-1beta expression and noradrenaline content was observed. The dissociation between hypothalamic cytokine gene expression and noradrenergic neuronal activity, the lack of sustained stimulation of the stress axes and the loss of sympathetic signals in the joints indicate that the immune-neuro-endocrine communication is disrupted during experimental arthritis, thus producing chronification of arthritis (27).

The alteration in immune-neuro-endocrine communication plays a role in osteo-articular damage. In an interesting study, LEW/N rats were selected as high-active (HA) and low-active (LA) animals according to neuro-endocrine and immune reactivity. During adjuvant-arthritis (AA), no differences in the severity of inflammation and corticosterone response were observed. However, LA rats had more osteoporosis, periostal new bone formation, and bone destruction than HA rats as determined by radiographs taken on day 30. Splenocytes of LA rats had lower production of IL-10 and IFN gamma. Histological examination revealed more intense factor VIII staining in arthritic joints of LA animals that indicates synovial angiogenesis with high plasma VEGF (an important angiogenic factor). Expression of RANKL, a crucial factor promoting bone resorption, was also higher in joints of LA animals. Lower production of bone-protective cytokines and a higher rate of angiogenesis leading to more synovial proliferation may be responsible for the severe joint destruction in LA animals (28)

In RA patients, cortisol levels similar to control subjects have been found, even in the presence of high pro-inflammatory cytokine levels, which represent an inefficient response in this pro-inflammatory environment from the early stages of the disease (29, 30). Subsequent studies showed that after experiencing physical or psychological stress, patients with RA have low cortisol levels due to defects in the HPA axis. Other studies showed that the circadian rhythm of cortisol is lost in patients with active RA, and the HPA axis response to surgical stress is impaired by defects in the expression of CRH (3).

These disruptions are mediated by hypothalamic and pituitary hormones and some pro-inflammatory cytokines such as IL-6 and TNF-α. In fact, it was demonstrated that anti-TNF- α therapy prevents abnormalities of the HPA axis in patients with RA and reduces fatigue, indicating that pro-inflammatory cytokines alter CNS function (31-33). A recent study showed that patients with good response to anti-TNF-α treatment also presented an increase in serum cortisol levels in contrast to the patients with poor response. This study demonstrated for the first time that, inflammation induced by TNF-α interferes with HPA axis integrity in humans and correlates with treatment response. Determination of plasmatic cortisol may be a sensitive marker to predict response to anti TNF-α treatment (34).

Hypothalamus-pituitary-gonadal axis (HPG)

In patients with RA, an abnormal metabolism of sex hormones has been demonstrated systemically and locally (in synovial fluid) starting from the early stages of the disease. While low serum concentrations of T, dehydrotestosterone, DHEA, and dehydroepiandrosterone sulfate (DHEAS) have been found, E levels are normal or even high (35). From the start of treatment with disease modifying antirheumatic drugs, a recent study evaluated changes in the HPG axis in men with early RA over a 2-year period. Men with RA, including patients older than 50, had mean T levels that were lower than the controls had early in the course of the disease which suggested mild hypogonadotropic hypogonadism. In patients who responded to treatment, T levels increased significantly. A decrease in DAS28 during the 2 year follow-up significantly correlated with increased T levels (36). A subset of glucocorticoid-naïve premenopausal females with RA had a relative hypocompetence of adrenocortical function with low levels of basal cortisol and DHEAS as well as alterations in adrenal synthetic pathways or deficiencies in steroidogenesis (37).

Furthermore, an accelerated conversion of androgenic precursors to 17-β-estradiol has been suggested. This imbalance in E metabolism results in an elevation of 16α-hydroxiestrone/4-hydroxiestradiol metabolites (38). These hydroxilated E metabolites have the ability to stimulate monocyte proliferation and growth and to play a role in synovial hyperplasia (39). These disturbances may be attributed to the fact that some pro-inflammatory cytokines (i.e., TNF-α, IL-1, IL-6) stimulate aromatase activity. This partially explains the abnormal peripheral synthesis of E in RA and their higher availability in the synovial fluid (40). This imbalance leads to a decrease in the increased androgen to estrogen conversion in the synovial fluid of RA patients and thus indicates the existence of a proinflammatory hormonal environment (41,42).

In addition, synovial cells express E receptors that correlate positively with synovial secretion of IL-6 and IL-8 (43). Therapeutic blocking of TNF-α has demonstrated a benefit because it increases DHEAS levels, indicating that biological therapy may improve the HPG axis function (44).

Autonomic nervous system (ANS)

The studies of the autonomic nervous system in patients with autoimmune diseases have shown conflicting results because symptoms of autonomic dysfunction are nonspecific and extremely varied. Parasympathetic and sympathetic dysfunction has been detected in 24% and 100% respectively of patients with AD. Cardiovascular autonomic dysfunction is the most common type of autonomic disturbance in patients with ARD (45).

Several abnormalities of ANS or the peripheral nervous system occur at different stages of RA (46). Activation of the sympathetic nervous system (SNS) and the parasympathetic nervous system in RA occurs parallel to HPA axis activation and to cytokine, hormone, and neuro-transmitter expression. This hypothesis is based on experimental models. In the arthritis model induced by type II collagen, it has been found that a transient increase in cortisol is followed by a rise in adrenaline levels and hypothalamic overexpression of IL-1β and IL-6 during the arthritis induction phase. The symptomatic phase showed an increase in hypothalamic noradrenaline followed by loss of noradrenergic fibers in joints. These findings indicate a disruption of immune-neuro-endocrine communication and noradrenergic activity during experimental arthritis (47).

In patients with juvenile idiopathic arthritis (JIA), an increase in SNS tone has been found. This data shows that the function of the ANS patients with JIA has been altered. The alteration is associated with a higher central noradrenergic outflow that presumably leads to increased vasoconstriction and results in a decreased response to an orthostatic stressor (48). Patients with early RA have high cardiac SNS activity while the parasympathetic activity is normal. These results suggest that inflammatory stress is responsible for these alterations in patients with RA (49). Tissue with inflammation from RA patients had less sympathetic nervous fiber in comparison with patients with osteoarthritis or trauma. Sensory nerve fibers contain two major neuropeptides, substance P (SP), and calcitonin gene-related peptide (CGRP). The pro-inflammatory role of SP is known, while CGRP has anti-inflammatory activities. In RA, there is an increase in sensory fibers and SP and a decrease in sensory fibers for the peptide related to the calcitonin gene (CGRP) in comparison with osteoarthritis (50,51). Some markers of SNS activity in patients with RA have been identified. Semaphorin 3C, a factor directed against sympathetic nerve fibers seems to be the main one responsible for reduction of tissue innervations in RA (52). Soluble neuropilin, a nerve repellent receptor, is another factor responsible for the disappearance of sympathetic nerve fibers soon after the beginning of inflammation (53). There are elevated levels of adrenal chromographin A in RA patients (54). Activation of SNS has pro- and anti-inflammatory effects, depending on the stage of the inflammatory process and the stimulated receptors. The acute phase of RA is characterized by β-adrenoreceptor stimulation with pro-inflammatory effects, whereas the chronic stage discloses stimulation of α-adrenoreceptors with anti-inflammatory effects (55).

The increased activation of ANS negatively influences the RA course by participating in the increase in cardiovascular risk observed in RA patients. Experimental evidence indicates that the SNS is critically influenced, at both the central and peripheral level, by the factors regulating vascular function such as nitric oxide, reactive oxygen species, endothelin, and the renin-angiotensin system. Additionally, there is indirect evidence of a reciprocal relationship between endothelial function and SNS activity (56).

Early detection of SNS hyperactivity in RA patients may be helpful in preventing these complications. In this regard, neuropeptide Y (NPY), an excellent indicator of sympathetic activity, is elevated in patients with RA and SLE and could be used as a sympathetic hyperactivity marker (57). However, another study found that NPY levels did not differ between RA and the control group or in heart rate variability parameters considered to reflect sympathetic activity (58). In any case, the HPA axis and the SNS act synergistically in many ways.

An association between variants in the gene encoding the beta 2-adrenergic receptor (ADRB2) and RA have been demonstrated. In contrast, no association was found between JIA and alleles, genotypes, or haplotypes of ADRB2, especially in the case of the haplotype R16/Q27. These observations suggest that JIA and RA have a specific genetic association (59).

In summary, in patients with RA and other chronic inflammatory diseases, an ANS activity with increased sympathetic nervous tone is reported. The reason for this high sympathetic activity is probably the inefficient cortisol production in RA and other types of chronic arthritis (60).

Prolactin-growth hormone system (PRL-GH)

Initial studies in humans did not show abnormalities in PRL and GH levels in RA patients. However, PRL bioactivity, using NB2 cells from mice lymphoma was significantly lower in comparison with healthy controls matched by age and gender which suggests a PRL deficiency in RA (61). The first evidence of a PRL role in arthritis was found in children who had juvenile chronic arthritis with positive antinuclear antibodies, and who had a significant PRL increase in direct correlation with IL-6 serum levels (62,63). However, studies in RA patients have not shown a consistently higher PRL. A correlation among higher PRL levels, clinical activity, and macrophage activity indicators such as macrophage inflammatory protein-1-alpha (MIP-1α) (64) was found. An interesting study in males with RA found that high PRL levels were associated with long-term RA and a poor functional classification (65). These findings have been confirmed, and they are associated with RA activity as well (66). PRL levels in synovial fluid from RA patients are similar to those of patients with osteoarthritis, but GH and other gonadal hormones and neuropeptides were also elevated in the synovial fluid of both groups of patients suggesting that these hormones have a local pro-inflammatory role (67). In summary, one third of the RA patients have been found to present basal HPRL indicating that this hormone plays a role in the inflammatory process of RA (17).

Hormonal functional studies allow us to analyze a normal or abnormal pituitary response. In RA, the results of these studies are controversial. One of the first studies demonstrated that patients with RA developed HPRL and a dissociation of the ACTH response after stimulation with TRH and CRH respectively independently of the clinical activity (68). This alteration is more evident in HLA-DR4+ patients than in HLA-DR4- patients and this suggests an alteration of HPA and PRL release in post-menopausal women with RA (69). Another study demonstrated that RA patients secreted excessive amounts of PRL during a stressful situation, e.g., surgery in comparison with chronic osteomyelitis patients. High PRL levels may contribute to disease activity by increasing the inflammatory/immune process regardless of genetic factors (70). However, in patients with a recent RA diagnosis and without steroid treatment or remission induction drugs, there was no difference between patients and controls in ACTH, cortisol, PRL, and TSH serum concentrations before and after stimulation with their respective hypothalamic releasing hormones (71). These discrepancies may be due to multiple factors including the hour at which the study was done. Studies done in the afternoon are not suitable since pro-inflammatory cytokine release is lower and the rheumatoid factor in early RA is absent; hence, this can suggests a diagnosis other than RA. Other studies employed an insulin induced hypoglycemia stress test in order to measure cortisol, TRH, and PRL in active RA patients without steroids and demonstrate an HPA axis dysfunction and normal PRL release (72). Another investigation found that PRL and GH responses to hypoglycemia stress were similar in RA patients under treatment and controls (73) although the receiver operating characteristic (ROC) curve analysis of PRL secretion was minor in RA patients. Other authors using the same stimulus found a lower PRL response in patients with non-treated active RA which normalized after conventional RA treatment. These findings suggest that RA activity and/or conventional treatment affect the regulation of central secretion of PRL (74). An analysis of this controversy indicates that there are patients who are highly responsive to the stress test and others that are not. Moreover, RA activity with the release of pro-inflammatory cytokines that modulate PRL response seems to be more important for assessing the response to hypoglycemia than for treatment (75).

In this regard, a study was done to evaluate the relationship between the level of leptin, PRL, IL-4, and IL-5 and the activity of RA and SLE. The authors concluded that while leptin cannot be used to evaluate disease activity in these entities, PRL can be utilized to assess disease activity in RA and SLE (76).

Despite these controversies, PRL receptors were identified in synovial cells and lymphocytes which infiltrate the synovial membrane with experimental evidence that T cells and fibroblasts of patients with RA may produce PRL. PRL increases synovial proliferation and the production of pro-inflammatory cytokines and metalloproteases. Bromocriptine (BRC) reduces not only PRL production but also pro-inflammatory cytokines and collagenases by RA synovial cells (77).

In support of these findings, a recent study evaluated the levels of PRL in the serum and synovial fluid of patients with RA and OA. The PRL that was found in the serum and synovial fluid was significantly higher in RA than in OA. The PRL synovial fluid correlated significantly with the RA activity and serum PRL levels with the total Larsen score. The increase in PRL may play a role in disease severity and the joint damage of patients with RA (78).

Non-controlled studies suggest clinical improvement of the disease with the use of BRC (79). Treatment of RA with BRC induced a significant decrease in the immune function in vitro, and these changes correlated with an improvement in activity parameters of RA (80). A case of RA refractory to conventional treatment improved with cabergolide, a PRL antagonist indicated for coincident HPRL treatment (81).

The PRL gene is located in close proximity to the HLA region on the short arm of chromosome 6. In this regard, it has been hypothesized that the associations between DR4 and reproductive risk factors in RA are due to linkage imbalance between DR4 and an abnormally regulated PRL gene polymorphism (82). In fact, there is linkage disequilibrium between HLA-DRB1 disease susceptibility alleles and microsatellite markers close to the PRL gene among women with RA and SLE (83). In contrast, none of the polymorphisms of neuroendocrine genes, including PRL gene, showed any statistically significant associations with JIA (84).

The PRL 1149 T (minor) allele decreases PRL expression and may be associated with AD. In order to determine the role of the PRL -1149 G/T polymorphism (rs1341239) in RA susceptibility, the association between PRL -1149 G/T and RA risk was examined in 3,405 RA cases and 4,111 controls. The results of this study indicate a possible association between the PRL -1149 T allele and decreased RA risk (85).

In conclusion, there is a growing body of evidence demonstrating an intriguing link between PRL and RA in experimental models and humans. New studies are necessary in order to find the PRL action mechanisms in RA and to test the efficacy of PRL antagonists in this disease (86).

Hypothalamus-pituitary-adrenal axis

The existence of a defect in the HPA axis in SLE has been shown in murine and human SLE. The MRL/lpr murine model presents a significant decrease in cortisol levels after stimulation with recombinant IL-1 (87). HPA axis studies are scarce and are influenced by treatment. Patients with SLE and moderate activity have decreased cortisol levels with normal ACTH both basal and after stimulation. Functional studies of responses to induced hypoglycemia demonstrated decreased cortisol in women with active SLE (88). In contrast, a recent study did not show alterations in the HPA axis in SLE patients without treatment. These contradictions may be due to the small sample size and the different methods employed (89). However the overall analysis of ACTH, androstenedione, cortisol, and DHEAS before and during a CRH test in patients with moderately active SLE undergoing low dose long-term glucocorticoid therapy showed marked adrenal insufficiency and a shift in steroidogenesis to cortisol in these patients but a completely normal pituitary function. This may depend in part on prior long-term glucocorticoid therapy and changes in steroidogenesis due to cytokines. Further longitudinal studies on untreated patients are necessary to investigate the endocrine-immune interplay and its consequences during the course of SLE (90).

Hypothalamus-pituitary-gonadal axis

The major risk factor for developing SLE is being young and female. Menses and pregnancy may cause an SLE flare. These observations strongly suggest a role for E and A in gender bias and in immunological and clinical manifestations of human SLE. In fact, abnormalities in both E and A metabolism in SLE patients have been demonstrated. Increased hydroxylation of estradiol at c-16 (potent E metabolite) was found in both males and females with SLE when compared to normal subjects. The conversion of upstream A precursors to 17β-estradiol is accelerated in RA and SLE patients (91,92). Specific abnormalities of A metabolism associated with SLE may contribute in some way to morbidity and mortality. The reduction of serum concentrations of DHEAS, T, and P in both male and female RA and SLE patients strongly supports the existence of accelerated peripheral metabolic conversion of upstream A precursors to 17β-estradiol. These data may have implications for future therapeutic regimens based on male hormone replacement (93).

A recent study utilized gene profiles of activated T cells from females with SLE and healthy controls which showed that E up-regulated six pathways in SLE T cells including interferon-alpha signaling. These results indicate that E alters signaling pathways in activated SLE T cells and contributes to SLE onset and disease pathogenesis (94). The ovarian function in SLE patients with active disease before treatment with high doses of glucocorticoids and cytotoxic agents was studied. Menstrual cycle disorders with oligomenorrhea were observed in 54% of SLE patients. The hormonal studies showed decreased P level, reduced E2 concentration, and increased levels of LH, FSH, and PRL. Menstrual period disorders, decreased P levels, and HPRL were found to be significantly related to a high SLEDAI score. SLE women may be considered a group that is at risk of developing an altered ovarian function (95). Menstrual disturbances are frequent and may be associated with pituitary dysfunction, which would confirm a decrease in P production. The follicular reserve seems to be low regardless of intravenous cyclophosphamide treatment (96).

The origin of altered ovarian function in SLE is diverse: SLE activity, organ damage, and immunosuppressive treatment. Ovarian necrotizing vasculitis is a rare complication described in SLE which may be another cause of gonadal dysfunction (97).

Sex hormone-binding globulin (SHBG) regulates the bioavailability of sex hormones to target tissues. In this regard, the distribution of the SHBG functional polymorphism Asp327Asn (rs6259) was analyzed in SLE patients and controls in a Polish population. A higher risk SHBG327Asn variant was found to lead to the development of SLE. SHBG has a much higher affinity for T than E, and the SHBG327Asn variant displays lower E clearance. The opposing effects of E and T on the immune system and on the imbalance in the levels of these hormones in SLE patients can be enhanced by the SHBG327Asn protein variant (98).

There is a remarkably low number of male patients with SLE. However, male SLE patients have severe renal involvement and they need treatment with cyclophosphamide and other drugs that may affect testicular function. Recently, testicular function in SLE has been analyzed. This study identified a high frequency of testicular Sertoli cell dysfunction in male SLE associated with semen abnormalities determined by urological evaluation, testicular Doppler ultrasound, hormone profile, and anti-sperm antibodies. In addition, inhibin B levels, a heterodimeric glycoprotein hormone produced almost exclusively by testicular Sertoli cells, was lower in SLE patients (99). Note that male patients with untreated hypogonadism, including Klinefelter’s syndrome, have a greater risk of developing autoimmune rheumatic diseases such as ankylosing spondylitis, SLE, RA, etc. (100).

In conclusion, female and male patients with SLE have alterations of the HPG axis as a consequence of clinically and immunologically active disease, immunosuppressive treatment, and genetic predisposition. The interactions between gonadal hormones and the immune system support the hypothesis that an abnormal immune neuroendocrine alteration may be the cause of clinical expression of AD.

Prolactin-growth hormone system (PRL-GH)

The relationship between PRL and the immune system has been demonstrated and created a new vision of immunoendocrinology. PRL is secreted not only by the anterior pituitary gland but also by many extrapituitary sites including the immune cells. The main function of PRL is to regulate the growth and differentiation of the mammary gland and the ovary and to act on the innate and adaptive immune response. HPRL has been described in both non-organ-specific ADs (e.g., SLE, RA, systemic sclerosis, psoriatic arthritis) and organ-specific ADs (e.g., celiac disease, type 1 diabetes mellitus, Addison’s disease, and autoimmune thyroid diseases). PRL increases the synthesis of IFN-gamma and IL-2 by Th1 lymphocytes. Moreover, PRL activates Th2 lymphocytes with autoantibody production (101,102). Experimental models and a few controlled studies of dopamine agonist treatment in humans with SLE support the potential efficacy of such agents even during pregnancy and postpartum. PRL is an integral member of the immune neuroendocrinology network. The presence of PRL mRNA in human lymphoid tissue implies that locally synthesized PRL may play a critical role in immunocompetence. PRL-receptors (PRL-R) are distributed throughout the immune system and are included as members of the cytokine receptor superfamily. PRL-R signal transduction is mediated by a complex array of signaling molecules of which the JAK2, Stat1, and Stat5 pathways have been well studied. PRL is now considered a cytokine based on both molecular and functional evidence (103-105). Structurally, PRL-R is related to the cytokine/hematopoetin family which includes the growth hormone (GH), the granulocyte-macrophage colony stimulating factor (GM-CSF), the erythropoietin, and the interleukin (IL)-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-9, IL-13, and IL-15 receptors. However, the effects of PRL depend on the local production by immunocompetent cells and its concentration as well as other hormonal factors and cytokines (106).

With respect to the different PRL actions, a recent study in mice susceptible to lupus found that HPRL accelerated the disease and increased the absolute numbers of T1 and T3 B cells but not of mature B cells. This suggested a primary effect of PRL on the early stages of B cell maturation in the spleen and a role for PRL in B cell differentiation, in which it contributed to SLE onset (107).

During the acute phase response (infection and various forms of injury), the adaptive immune response is suppressed and the innate immune function is amplified. The HPA axis stimulates natural immunity and suppressor/regulatory T cells, which down-regulate the adaptive immune system. Adaptive immunocompetence is maintained by vasopressin (VP), GH, and PRL. VP controls adaptive immune function and stimulates ACTH and PRL secretion (108).

Estrogens have long been regarded as major physiological activators of PRL synthesis and lactotroph proliferation and have been implicated in the pathophysiology of HPRL and prolactinomas. E2 stimulates PRL gene expression through binding E receptors. E2 and TNF-α have synergistic activation of the human PRL promoter. This effect is mediated by nuclear factor kappa B (NF-kappa β) thus suggesting a novel, promoter-specific signaling interaction between E and TNF-α for PRL regulation in vivo (109).

Early reports showed HPRL in SLE patients. The first study was done on male SLE patients without treatment. These patients had high levels of PRL both before and after intravenous administration of LH–RH stimulation associated with increased estrone and decreased T, and DHEA levels compared to healthy controls (110). Pregnant SLE patients also had HPRL as did healthy pregnant women and RA patients, which is associated with E and T decrease, especially in active SLE (111). An open study demonstrated significant HPRL in SLE patients in comparison to normal controls and other ARD with a direct correlation with clinical and serological activity suggesting that PRL plays a role in SLE pathogenesis (112). HPRL has been seen in SLE patients, but whether HPRL is associated with SLE activity is still controversial. These discrepancies may be explained by multiple factors including heterogeneous groups of patients, variability of the SLE activity index, treatments, circadian PRL rhythm, and different isoforms of PRL and anti-PRL antibodies in the sera of SLE patients (113). Despite this controversy, HPRL and elevated circulating IL-6 were reported in SLE patients with active lupus nephritis or neuropsychiatric involvement. These findings support the hypothesis that PRL and IL-6 are accumulated in the inflamed organs and act as stimuli for T and B lymphocytes to infiltrate the tissue. It is noteworthy that PRL-R has been described in the choroid plexus, hypothalamus, and other areas of the central nervous system. In addition, at 7.5–14 weeks of gestation, PRL R is expressed chiefly in the human fetal kidney (113-116). The high concentrations of PRL and IL-6 in urine and cerebral spinal fluid may be the consequence of local synthesis of both mediators. A subset (30–40%) of SLE patients with HPRL has detectable anti-PRL antibodies with fewer clinical and serological manifestations. In contrast, SLE patients without macroprolactinemia (PRL-IgG complex) had significantly higher levels of disease activity compared to patients with macroprolactinemia. PRL-IgG complex was produced by 23-kd non glycosylated PRL that bound to IgG, and it was fully active in vitro. Therefore, the absence of symptoms of HPRL or lower levels of lupus activity in patients with anti-PRL antibodies is not explained by the lower bioactivity of the complex. Delayed clearance of the PRL-IgG complex may account for increased serum levels of PRL in SLE patients with anti-PRL autoantibodies (117).

The association of PRL with disease activity has been analyzed over the last 10 years. In 2001, three consecutive studies of independent centers demonstrated the association among PRL levels and disease activity, malar rash, CNS involvement, fatigue, fever, and renal involvement in 397 SLE patients. Additionally, anti-dsDNA, anti cardiolipin antibody titers, elevated erythrocyte sedimentation rate, anemia, low levels of C3 were associated with HPRL. Serum PRL concentrations were determined by immunoradiometric (IRMA) and biological assays which evaluate Nb2 lymphoma cell proliferation. Therefore, high serum PRL levels are an important independent factor related to lupus activity (118-120). Recent studies confirmed these findings, and a significant correlation between PRL levels, clinical disease, and serological activity of SLE was found. In addition, menstrual cycle disorders, decreased P level, and HPRL were found to be significantly associated with a high SLEDAI score before treatment with cyclophosphamide and high doses of GC (121, 95). According to meta-analysis of the role of PRL in SLE, a significant increase in PRL concentration in SLE patients has been demonstrated. Serum PRL levels have been positively associated with lupus disease activity. Abnormally high PRL levels during pregnancy in SLE also correlate with disease activity (122).

In vitro studies from human cells support the association of PRL and SLE. Lymphocyte derived PRL has been shown to be higher in patients with SLE compared to normal subjects (123,124). Jacobi et al. (125) found that physiological concentrations of PRL (20 ng/ml) induced IgG production more effectively than high concentrations of PRL (100 ng/ml) in lymphocytes from active SLE patients. As a result, PRL-induced IgG production was associated with SLE activity. Chavez-Rueda et al. (126) showed that early activation of T and B lymphocytes from SLE patients by PRL was associated with disease activity and PRL production by lymphocytes. The addition of anti-PRL antibodies resulted in a reduction of CD69 and CD154 expression. PRL impairs the negative selection of autoreactive B lymphocytes that occurs during B cell maturation into fully functional B cells. PRL has an anti-apoptotic effect and enhances proliferative response to antigens and mitogens and the production of immunoglobulins and autoantibodies (127). Controlled clinical trials demonstrated that treatment with BRC reduced the expected number of lupus flares and improved anxiety and distress and thus supported the role of PRL in disease activity. BRC is currently considered an unproven therapy for SLE, and its use is entirely experimental (128). HPRL in patients with lupus activity at onset and after 6 months of conventional treatment was associated with both SLE disease activity and remission induced by conventional treatment (129). Acute phase proteins, PRL, and disease activity in SLE patients have been evaluated during quinagolide therapy at onset and after 3 months of treatment. A significant decrease in the SLEDAI score, IL-6, and PRL levels was revealed after 3 months of therapy confirming the hypothesis that quinagolide may become a valuable and safe drug for SLE treatment (130). HPRL, hyperferritinemia, and hypovitaminosis D have different immunological implications in the pathogenesis of ADs, including SLE. It has been recently suggested that, in addition to therapy for HPRL with dopamine agonists, preventive treatment with vitamin D may be considered in certain cases (131). Due to deficiency of dendritic cell functions and abnormal PRL secretion in SLE, another possible participation of PRL in the pathogenesis of SLE is the interaction of PRL and dendritic cells. The interaction skews their function from antigen presentation to a pro-inflammatory phenotype with high INF-α production (132). There are patients with SLE in whom no secondary cause for HPRL can be found. Defects in peptidergic modulators and dopamine metabolism may, in part, explain the HPRL in SLE. Lymphocytes in active SLE have an increased production of PRL. Impaired hypothalamic function has been associated with HPRL in SLE (133,134). Moreover, STAT5 signaling is very important in the regulation of the immune response in SLE. Basal activation of STAT signaling and a reduced response to the cytokines IFNα and IL6 were observed in the peripheral blood of SLE patients. The phosphorylation of STAT5 is associated with cytokines such as PRL, IL2, G-CSF, and IFNγ (135).

In conclusion, experimental models, in vitro studies, and clinical observations have shown an intriguing link between PRL, autoimmunity, SLE, and other ADs. HPRL is associated with active disease and organ involvement in SLE patients. This knowledge might have important implications for treatment. Further studies are necessary to confirm the efficacy and safety of PRL lowering agents as therapy for SLE.

Autonomic nervous system (ANS)

The manifestations of autonomic nervous system (ANS) dysfunction in ARD have been the subject of many studies. In one case, a study found that while RA and SLE patients had depressed heart rate variability compared to controls, the QTc interval was more prolonged in patients with SLE. In this study, autonomic dysfunction in SLE was predominantly sympathetic while in RA vagal predominance was evident (136).

In patients with moderately active SLE, an altered ANS response which was more pronounced in comparison to healthy subjects was found during the hCRH stress test. The authors concluded that a hypersympathetic reaction may lead to a greater risk of cardiovascular diseases in these patients (137).

The autonomic neural control of cardiovascular and immune functions involves a number of autonomic neuropeptides such as neuropeptide Y (NPY) and vasoactive intestinal peptide (VIP), which participates in the orchestration of cytokines and exerts modulatory effects on immune cells. In a study of adolescents with SLE and juvenile idiopathic arthritis (JIA) patients, both groups had significantly lower serum NPY and VIP with an association between cardiovascular autonomic neuropathy and disease manifestations including activity. This finding may denote hypofunction of sympathetic and parasympathetic divisions of ANS in lupus (138).

In contrast, another study found an increase in sympathetic outflow measured by NPY and a rise in the NPY/ACTH ratio in SLE patients which suggests that low levels of cortisol in relation to neurotransmitters may be pro-inflammatory because cooperative anti-inflammatory coupling of the two endogenous response axes is missing (57). Further studies are needed to elucidate the prognostic significance and clinical implications of impaired autonomic functions as well as the immune neuroendocrine network in patients with SLE (e.g., cardiovascular risk).

Hypothalamus-pituitary-adrenal axis

The use of corticosteroids as the cornerstone for the treatment of uveitis strongly suggests an abnormal immune-neuroendocrine interaction in these localized AD. CRH is a major regulator of the HPA and the main coordinator of the stress response. During inflammatory stress, the cytokines TNF α, IL-1, and IL-6 stimulate hypothalamic CRH and/or vasopressin secretion as a way of preventing an overreacting inflammation. This response has been observed in experimental models of inflammatory/immune diseases including uveitis induced by R16 peptide in Lewis rats (140).

In autoimmune uveitis, especially that associated with ADs, there are some studies that suggest an inhibition of the HPA axis characterized by significantly low levels of cortisol and HPRL as well as high levels of pro-inflammatory cytokines (i.e., TNF-α, IL-1, and IL-6) (141,142). The serum levels of cortisol and glycosaminoglycans (GAG) were studied in patients with uveitis in ARD. In acute uveitis, the mean serum cortisol level was significantly lower than that in infectious and idiopathic uveitis. In ARD, remission in uveitis was associated with an increase in the serum levels of cortisol. A poor course was observed when the low content of cortisol remained or decreased still further. GAG exchange in uveitis patients with RD changed: there was an inverse correlation between the level of hyaluronic acid and cortisol. The same correlation between GAG and chondroitin sulfate was found in males (143).

Hypothalamus-pituitary-gonadal axis

Gender and sex hormones influence ocular diseases including uveitis and retinal disease as well as eye circulation, and optic nerve anatomy. Sex-based differences in ocular disease should be considered (144).

Pregnancy and postpartum period are associated with uveitis activity. In this regard, a prospective, observational case study was done. Four pregnant women in their first trimester with chronic non-infectious uveitis were followed monthly until 6 months after delivery. Serum female hormones (i.e., E, P, PRL) and cytokines (e.g., IL-2, IL-4, IL-5, IL-6, IL-10, IFN-γ, and TGF-β) were measured. Uveitis activity decreased after the first trimester but flared in the early postpartum period. Serum female hormones, after being highly elevated during pregnancy, drastically dropped in the postpartum. Cytokine levels, except TGF-β, were mostly undetectable suggesting that female hormones and TGF-β may contribute to the activity of uveitis during pregnancy and the postpartum period (145).

The relationship between hormones and the immune system in autoimmune uveitis is more evident in experimental autoimmune uveitis (EAU). The effect of E, P, and T in EAU was studied. In female rats, T decreased significantly, E was slightly enhanced, but P or E + P did not affect EAU. A correlation with the ocular levels of Th1 (IFN-γ) and Th2 (IL-10) cytokine messengers was found thereby suggesting that sex hormones may affect EAU by inducing changes in the cytokine balance. Sex hormone therapy could be considered an adjunct to anti-inflammatory agents used to treat autoimmune uveitis in humans (146). In this regard, the endotoxin-induced uveitis (EIU) and the effects of E were studied in adult male, female, and ovariectomized female Lewis rats. The endotoxin injection produced cell infiltration, which was more marked in male than in female rats and in rats with ovariectomy. E receptor was found in the endothelium and iris-ciliary body, and E-selectin and IL-6 gene expressions were higher. Note that treatment with 17-beta-estradiol significantly reduced cell infiltration and gene expression in male and ovariectomized female rats. The down-modulation of these inflammatory genes by E may contribute to the reduction in cell infiltration in acute anterior uveitis (147).

Prolactin-growth hormone system (PRL-GH)

Prolactin (PRL) may play a role in the regulation of humoral and cell-mediated immune responses. On the basis of these observations, in 1991, PRL levels were measured in the serum and aqueous humor of 28 patients with cataract or anterior uveitis with concomitant cataract. Intraocular concentrations were significantly higher in uveitis patients. This was the first study measuring PRL concentrations in human aqueous humor (148). Previous studies suggested that experimental autoimmune uveitis as well as patients with uveitis, whether or not associated with spondyloarthritis, may respond to BRC treatment. (101,149). Serum levels of the hormone melatonin (MEL), PRL, and IL-2 were measured in 100 patients with uveitis and matched with healthy blood donors. MEL was reduced significantly in patients with iritis and iridocyclitis and significantly elevated in those with intermediate uveitis, chorioretinitis, and panuveitis. PRL was significantly reduced in patients with intermediate uveitis. IL-2 was reduced to about 50% of control values in all groups of patients. The results suggest a possible immune neuroendocrine interaction in uveitis patients (150). In contrast, a recent study evaluated basal serum PRL levels in patients with HLA-B27-associated uveitis in a prospective, nonrandomized comparative trial. Thirty-three patients with HLA-B27- associated uveitis and 30 age- and sex-matched healthy control subjects were included. PRL serum levels were significantly higher in patients vs. controls. However, no correlation was found between PRL levels, systemic treatment, and disease activity. These results suggest the role of serum prolactin levels in HLA-B27-associated uveitis pathogenesis. In a previous study, we found HPRL in patients with Reiter’s syndrome. The frequency of conjunctivitis, urethritis, dysentery, and uveitis was higher in patients with HPRL than in normoprolactinemic patients with Reiter’s syndrome (152).

In conclusion: uveitis is a group of inflammatory eye diseases with unspecified etiology. There is evidence that uveitis is an autoimmune disease mediated by T cells, but the triggering mechanism is unknown. A possible cause of such conditions is the disturbance at the immune neuroendocrine system level with alterations on the HPA axis, the HG axis, PRL hormone, and cytokine release. Finally, a variety of evidence suggests that uveitis is a systemic disease rather than simply an ocular disorder.

Rheumatic diseases, cortisol, and TH1 cytokines

Cortisol, an adrenal steroid with anti-inflammatory functions, has a circadian rhythm with a maximum peak at 8 A.M. and nadir at midnight. In healthy subjects, it has been observed that bone resorption activity is higher between 5 to 7 A.M., because resorption is adjusted to cortisol circadian rhythm, TNF-α / IL-6, thus contributing to loss of bone mass (154). The circadian rhythm of cortisol in RA patients and low to moderate disease activity are not different from the rhythm of healthy subjects. However, as has been demonstrated, this rhythm may be altered in RA patients with high disease activity and thus lead to flattening of the hormone curve and the appearance of two peaks in the morning and afternoon (155). Although these patients have higher cortisol levels, they are insufficient in relation to the inflammation and, as a result, the altered circadian rhythm causes greater inflammation during the early hours of the morning (156).

Patients with apparently well-controlled RA may have debilitating symptoms such as morning stiffness, fatigue, and pain. The key to controlling these symptoms may be in understanding their pathophysiology. Nocturnal plasma levels of the pro-inflammatory cytokine IL-6 are elevated in patients with RA and correlate with levels of morning stiffness. Endogenous cortisol secreted during the night may be insufficient to counter the actions of IL-6 in these patients. Consistent with this hypothesis, the beneficial effects of glucocorticoids on morning stiffness are enhanced by their administration at 02:00 h compared to conventional administration around breakfast time though it is inconvenient for patients to have to awaken to take therapy. Modified-release prednisone has been developed to allow treatment to be taken at a convenient time (≈ 22:00 h) with programmed delivery of the glucocorticoid 4–6 h later, which is a more appropriate time. Assessment of cytokine and cortisol levels over 24 h before and 2 weeks after treatment with modified-release prednisone 5 mg/day has confirmed the hypothesis. Clinical studies in patients with RA have shown that modified-release prednisone at the same dose significantly reduced the duration of morning stiffness without affecting tolerability. Furthermore, the administration of glucocorticoid in accordance with the natural circadian rhythm may improve hypothalamic-pituitary-adrenal axis function (157).

According to these clinical findings, the molecular machinery responsible for the circadian timekeeping is perturbed in RA. A recent study investigated the expression of the molecular circadian clock in RA. Gene expression of thirteen clock genes was analyzed in the synovial membrane of RA and control osteoarthritis (OA) patients. The effect of pro-inflammatory stimulus on clock gene expression in synovial fibroblasts was studied using IL-6 and TNF-α. Gene expression analysis disclosed disconcerted circadian timekeeping and immunohistochemistry revealed a strong cytoplasmic localization of BMAL1, a transcription factor of circadian rhythm, in RA patients. Perturbed circadian timekeeping is at least in part inflammation independent and cell autonomous, because RA synovial fibroblasts display altered circadian expression of several clock components and perturbed circadian production of IL-6 and IL-1β after clock resetting. Throughout the experiments, ARNTL2 and NPAS2 appeared to be the most affected clock genes in human immune-inflammatory conditions. This study demonstrated that the molecular machinery controlling the circadian rhythm is disturbed in RA patients (153).

Participation of other hormones.

It was demonstrated that women with active RA showed a tendency towards GH insensitivity and a decrease in diurnal cortisol and DHEA in relation to their inflammatory state (158).

Melatonin, another hormone with a night circadian cycle produced by the pineal gland has the opposite effect of steroids. The elevated secretion in patients with RA seems to be an important factor for perpetuation and clinical circadian disease symptoms (159). In active RA, there is an imbalance favoring pro-inflammatory hormones (PRL and cytokines) which is responsible for the diurnal inflammatory rhythm of the disease (160). Furthermore, cytokines also have a circadian oscillation since IL-6, and TNF-α present maximum levels during the early morning hours. The IL-6 peak is at around 6 A.M. in healthy subjects with a decrease at 9 A.M. It is at 7 A.M. in RA patients and persistently rises until 11 A.M. thus influencing morning symptoms (161).

Perspectives of treatment

Reprogramming biological rhythms has recently gained a lot attention as a potent method to leverage homeostatic circadian controls to ultimately improve clinical outcomes. Elucidation of the intrinsic properties of such complex systems and optimization of intervention strategies require not only an accurate identification of the signaling pathways that mediate host responses but also a system-level description and evaluation (6).

Considering circadian rhythms and the impact of immunosuppressive schedule intake on efficacy, it would be of great interest to explore new therapeutic modalities such as those to neutralize cytokines, melatonin hormonal antagonists, new formulation of old drugs towards other release patterns and a major night inhibition of pro-inflammatory cytokines such as IL-6 and TNF-α, antagonists of IL-1 receptor. And so forth which of them are targeted at reducing RA symptoms during the early morning hours and improving quality of life for these patients. One of these new therapeutic agents, Tocilizumab, is a monoclonal antibody against IL-6 receptor. Clinical studies done of RA patients over the last few years have demonstrated that it is effective at improving symptoms and signs of RA. Moreover, inhibition of IL-6 may have positive effects on the functional status and radiological progression of the disease (161). Another interleukin expressed in the synovial membrane of RA patients is IL-17 that, with IL-1 and TNF-α, has a synergistic effect. IL-17 is a potent inductor of NF-kappaB. Anti-IL-17 therapy may be a new treatment option for bone destruction prevention along with anti-TNF-α and anti-IL-1 therapy (162). Regulation of transduction signals with small molecule inhibitors in rheumatic diseases, e.g., specific Jak3 inhibitors, spleen tyrosine-kinase (Syk) inhibitors, or mitogen activated kinase-like protein (MAPK) and endogen negative regulators of MAPK signals that have received recognition as inflammatory/immune response modulators with potential use in RA. Additional benefits seen in inflammatory disease treatment targeting CD80, IL-12/IL-23, AP-1 transcription factor, and cell activation receptor modulators such as cytokine receptors, toll-like receptors, and adenosine A3 receptors are being developed now (163). All of these potential treatments will probably modify the immune neuroendocrine response in order to stabilize the hormonal homeostasis.

The modulation of hormonal milieu has been a therapeutic option in ADs for several years. Clinical trials of DHEA at an oral dose of 200 mg/day for SLE patients have demonstrated an improvement in disease activity among female patients with mild to moderate lupus but to a lesser extent in cases of more severe disease. DHEA improves the ability to taper corticosteroids in patients with active lupus (164).

In relation to Danazol, a recent study reviewed the information on the use, effectiveness, and adverse effects of danazol in patients with SLE. The study ran from January 1950 to July 2009 with a total of 153 patients, including 2 prospective trials of 7 and 16 patients respectively and 1 randomized controlled trial of 40 patients. Danazol has been used successfully in the treatment of hematological manifestations of SLE such as thrombocytopenia, Evan’s syndrome, autoimmune hemolytic anemia, and a case of red cell aplasia. Thirteen patients responded to danazol after failing splenectomy. There is limited information on the use of danazol in nonhematological manifestations of SLE. Adverse effects were generally tolerable but high doses may produce undesirable side effects for female patients. Patients with primary antiphospholipid syndrome and RA were also treated with danazol with acceptable platelet counts within the first 4 weeks of danazol therapy that allowed the prednisone dosage to be tapered. (165-167).

Murine models of SLE demonstrate that PRL impairs the mechanisms of B cell tolerance induction (negative selection, receptor editing, and anergy), and therefore, it participates in the pathogenesis of autoimmunity. BRC, a drug that inhibits PRL secretion, restores the immune tolerance, and abrogates several immune effects of PRL (168). A double-blind, randomized, placebo-controlled study of BRC in SLE (mean follow-up 12.5 months) showed a decrease in the SLEDAI score and a significant reduction of flares/patient/month in the BRC group vs controls. Long term treatment with a low dose of BRC appears to be safe, and it is an effective means of decreasing SLE flares in these patients (169). Our pilot study suggests that BRC may play a role in the prevention of maternal-fetal complications such as premature rupture of membranes, preterm birth, and active disease (170).