Autonomic and immune dysfunction, and dysregulation of their cross-communication in autoimmune diseases, are hallmarks of many autoimmune diseases, including rheumatoid arthritis (RA) (1–4). In RA, chronic inflammation, “self” targeted arthritogenic T cells and autoantibodies drive the damage in the affected joints and visceral organs (5). Thus, altered innate and adaptive immunity are key mediators of the disease process (5, 6). Dysregulation of the immune system is coupled with hyperactivity of the sympathetic nervous system (SNS). Dysregulation of the SNS in the immune system contributes significantly to disease onset, and greater severity in RA and experimentally-induced arthritis (7, 8). The contribution of a dysregulated SNS and the altered cross-communication between SNS and the immune system in autoimmune diseases is complex. In adjuvant- and collagen-induced arthritis models [adjuvant-induced arthritis (AA) and CIA, respectively], the SNS exerts a pro-inflammatory influence during the inductive, asymptomatic phase, but suppresses the inflammatory and cell-mediated response after disease onset (9, 10). Thus, we still do not understand the complex and diverse roles of the SNS in regulating innate and adaptive immunity that mediate rheumatic disease mechanisms prior to and after disease onset. Growing evidence indicates that the SNS is an important target for developing disease amelioration, or even preventing disease onset, supporting the need to understand sympathetic-immune system cross-talk in RA (2).

We have shown disease promoting effects of sympathetic innervation of secondary lymphoid organs during the induction phase, a time of antigen processing, and disease enhancing effects in the effector phase of AA (11, 12). This finding was subsequently confirmed using the collagen-induced arthritis model of RA (13). In these studies, sympathetic nerves were destroyed by selective toxins (11, 13) or β₂-adrenergic receptor (AR) agonist and/or α-AR antagonist treatment (12) prior to the autoimmune inducing challenge or after disease onset. Treatment with a β₂-AR agonist and/or α-AR antagonist shifts the balance of pro- and anti-inflammatory cytokine production toward increased disease severity if administered prior to or reduced disease activity after disease onset, respectively. These findings indicate that changes in the nerves and receptors contribute to the opposite disease outcome effects of the SNS observed after challenge and disease onset. Consistent with this notion, we (14–16) and others (17) previously reported dramatic changes in the density and distribution of sympathetic nerves to discrete compartments in secondary immune organs in rodent models of RA, and in the affected joints of RA patients. These anatomical changes suggest that sympathetic nerves interact with various immune cell populations in different activational states and functional compartments. They also implicate a link between consequent changes in sympathetic-immune cell signaling with disease development and progression.

The SNS regulates the functions of immune cells that mediate rheumatic disease through the activation of α- and β₂-ARs expressed on their cell surfaces (10, 17–19). Sympathetic nerves target immune cells in secondary lymphoid organs (15, 16), and the immune cell infiltrates and tertiary lymph nodules that occur in the affected joints (20). In secondary lymphoid organs, neurotransmitters, predominantly norepinephrine released from sympathetic nerves, activate β₂-ARs expressed in T cells to inhibit IL-2 production, which subsequently suppresses lymphocyte proliferation required for clonal expansion [reviewed in Ref. (21)]. Activation of β₂-ARs also inhibit cellular and promote humoral immunity by regulating the phenotypic differentiation of CD4+ T helper (Th) cells in response to challenge with T-cell-dependent antigens (21). This occurs by norepinephrine activating Th0/Th1 cell β₂-ARs, which increase intracellular cAMP production. Cyclic AMP then inhibits IFN-γ production. The decrease in IFN-γ reduces the inhibitory cross-regulation of IFN-γ on IL-4 production, thus promoting IL-4 synthesis by Th2 cells. In this manner, the SNS provides a negative feedback mechanism to restore immune system homeostasis after antigen challenges that activate cellular immunity (21).

Recently, we found that activation of β₂-ARs no longer induces cAMP in immune cells in the spleen and lymph nodes that drain the arthritic hind-limbs in AA during severe disease (18). Altered β₂-AR responses occurred concomitantly with altered receptor-ligand affinity and lymphoid tissue changes in receptor phosphorylation. These findings predict that a negative feedback mechanism required to restore immune system homeostasis after adjuvant challenge is lost after disease onset. However, the impact of sympathetic regulation of IL-2, IFN-γ, and IL-4 production in disease-relevant secondary lymphoid organs has not been determined.

The SNS also time-dependently regulates pro- and anti-inflammatory cytokine production relative to the antigen challenge [reviewed in Ref. (21)]. After antigen challenge, norepinephrine released from sympathetic nerves at the site of challenge promotes TNF-α and suppresses IL-10 production by activating α-ARs expressed in macrophages to amplify the ensuing inflammatory response. Similarly, norepinephrine promotes TNF-α and suppresses IL-10 production in dendritic cells from secondary lymphoid tissue via α-ARs. In contrast, activation of β₂-AR in macrophages and dendritic cells suppresses TNF-α and promotes IL-10 production. The balance between the anti- and pro-inflammatory effects of the SNS is dependent upon the balance between α- and β₂-AR expression, intracellular signaling pathways they activate and the dynamic local environment that follows the arthritis inducing challenge (9, 12). In this manner, the SNS regulates the initiating and shutting off of inflammatory and innate immune responses. In animal models of RA, α-AR antagonists and β₂-AR agonists promote or inhibit joint inflammation if administered prior to or after disease onset, respectively. The effects of altered sympathetic innervation and changes in β₂-AR function on SNS regulation of pro- and anti-inflammatory cytokines seen previously in secondary lymphoid organs of arthritic rats have not been examined.

The purpose of this study was to determine how the SNS affects ex vivo cytokine production in secondary lymphoid organs during the effector phase of disease. Specifically, we examined the effect of AR selective drug treatments in (1) modulating IL-2 and proliferation, and (2) the balance between IFN-γ and IL-4 and between TNF-α and IL-10 after AA development. A specific β₂-AR agonist and an α-AR antagonist, alone and in combination [previously designated SH1293 (12)] were administered in vivo after disease onset. These treatments were used to determine the contribution of each receptor subtype in altering cytokine production and disease outcome. T cell and macrophage cytokines were measured ex vivo for each tissue collected from each treatment group. Cytokines with crucial roles in Th cell differentiation and clonal expansion or inflammation were selected for assessment: (1) immune cell production of IL-2, an important cytokine for development and differentiation of Th cells and proliferative responses required for clonal expansion (22); (2) IL-4 and IL-10, cytokines, which promote Th2 cell development and have anti-inflammatory functions; (3) IFN-γ and TNF-α, which promote Th1 cell development and which drive inflammation (23, 24).

In this study, we confirm and extend our previous findings of the disease-modifying effects of a β₂-AR agonist and/or α-AR antagonist treatment in AA after disease onset. Consistent with previous findings, TERB, PHEN, or PT dramatically ameliorates AA in male Lewis rats when treatment begins at disease onset (12). Our data extend these findings to show that all drug treatments suppressed proliferative responses in cell cultures from all lymphoid tissues examined. However, IL-2 production is not coupled with proliferation regardless of the adrenergic therapy. Immune cells in secondary lymphoid organs and peripheral blood are largely non-responsive to treatment with the β₂-AR agonist. These findings support that the SNS loses its ability to regulate immune cell functions via normal β-AR signal transduction in secondary lymphoid organs in AA and by extension RA. Importantly, the exceptions to these findings are the dramatic β₂-AR-mediated increase in spleen cell IL-10 and DLN cell IFN-γ production, respectively. Surprisingly, treatment with the α-AR antagonist produces greater changes in cytokine profiles than targeting β₂-ARs, suggesting a shift in the class of AR that predominantly regulates immune function during chronic disease. In sites not expected to elicit robust immune responses to CFA challenge (i.e., PBMCs and MLN cells), PHEN and PT suppresses IFN-γ and TNF-α cytokine production. In spleen cells, the major effect of PHEN or PT was to drive IL-10 and suppress TNF-α production. This cytokine profile is expected to suppress inflammation and drive Th2 responses, which are consistent with improved disease outcome of PHEN or PT treatment. In contrast to single drug treatments, PT increases IL-4 and IL-10 production in DLN cells, which is expected to suppress inflammation and promote Th2 immunity. This difference in cytokine profiles may account for PT having the greatest ameliorating effects on disease, particularly its bone sparing effects.

All adrenergic therapies reduce disease severity. Radiographs and scoring provide support for a direct and/or indirect effect of adrenergic drugs on all scoring components used to assess soft tissue swelling (i.e., inflammation) and bone destruction/remodeling in the arthritic hind-limbs. Direct effects of adrenergic agents have been reported for both cartilage and bone. Both chondrocytes and bone cells (osteoclasts and osteoblasts) express functional ARs (26, 27) and can alter their growth and resorption activity (28–31). Additionally, immune cells (i.e., macrophages, T, B, and NK cells) and their products (i.e., TNF-α, IL-1, IFN-γ, and IL-2) coordinate changes in bone cell function and are regulated by ARs (32–34). Our findings support direct and/or indirect effects of TERB, PHEN, and PT treatments in ameliorating inflammation and joint destruction, and indicate these treatments are disease-modifying, anti-rheumatic drugs (DMARDs). X-ray scores suggest that TERB was more effective at reducing swelling than either PHEN or PT based on soft tissue assessment, in agreement with other reports demonstrating β₂-AR-mediated anti-inflammatory effects (35–37). The β₂-AR-mediated decrease in inflammation may be mediated by inhibiting macrophage TNF-α production and/or reducing reactive oxygen species (35–37). Surprisingly, PT was much less effective in reducing soft tissue swelling than specifically targeting β₂-, or α-ARs. This finding suggests opposing actions of α- and β-ARs in the inflamed hind limbs. Whether this interaction involves multiple cell types and/or direct or indirect interplay between α- and β-ARs in the same cell types is unclear. Provocative new data (38, 39) indicate that β- and α-AR subtypes can form heterodimers that affect downstream signaling to regulate the inflammatory response. Similar research is warranted in our model to better understand β- and α-AR-mediated mechanisms regulating soft tissue swelling.

Immune cells cultured from lymphoid organs of CFA-challenged rats proliferate ex vivo without any additional immune challenge, an approach that better mimics the in vivo setting. Immune cell proliferation was most pronounced in DLNs and spleen. This finding is consistent with reports of persistent high antigen load in these tissues (40) and the significant involvement of both the DLNs and spleen in disease induction and immune dysregulation during the effector phase of the immune response. In response to CFA challenge, arthritogenic T cells that can transfer disease to naïve rats are generated in the spleen and DLNs (41). In agreement with this, we observed the greatest proliferative responses in DLNs and spleen cells, whereas proliferation was barely above background in PBMCs and MLN cells. TERB, PHEN, or PT treatment significantly reduced cellular proliferation in the spleen, DLN, and MLN, but had no effect on PBMC proliferation. This is consistent with the normalization of spleen weights after adrenergic treatments and with secondary lymphoid organs being sites of lymphocyte clonal expansion.

The reduced proliferative responses with TERB treatment are also consistent with other studies that demonstrate β₂-AR agonists and norepinephrine treatment reduce proliferation of antigen- and mitogen-challenged lymphocytes from secondary immune organs [reviewed in Ref. (21)]. Upon antigen challenge, T cell proliferation is promoted by the binding of IL-2 to the IL-2R expressed on T cells [reviewed in Ref. (21)]. Numerous studies have shown that norepinephrine/epinephrine decrease the proliferative responses in T cells by a β₂-AR-mediated reduction in IL-2 production following mitogen, anti-CD3 antibody, or antigen challenge (21). However, TERB treatment had no significant effects on IL-2 in our study, despite the β₂-AR-mediated reduction in proliferation. These findings indicate that IL-2 does not drive the lymphocyte proliferation observed in secondary lymphoid compartments during the effector phase of AA, and that TERB-induced decreases in lymphocyte proliferation are not mediated by IL-2. Reports that chronic secretion of TNF-α can cause IL-2 deficiency at sites of inflammation (42–44) is consistent with our high TNF levels, low IL-2 production and uncoupling of IL-2-induced proliferation. Moreover, our findings are consistent with T cell hyporesponsiveness to T cell receptor (TCR) engagement (42), with a switch from an IL-2-to-inflammatory-driven immune response that suppresses IL-2 gene transcription (44). Finally, the inability of TERB to reduce IL-2 production in spleen and DLN cells from AA rats is consistent with recent findings from our lab that β₂-ARs are uncoupled from cAMP (18). Others have reported that IL-2 decreases β-AR density, removing a negative control over cellular proliferation (45). Similarly, in certain cell populations, IL-1β can impair β₂-AR coupling with adenylate cyclase (46, 47). Future studies are needed to determine the mechanisms for the β₂-AR-mediated suppression of proliferative responses observed in secondary lymphoid organs in AA rats.

Surprisingly, treatment with the α-AR antagonist also reduced lymphocyte proliferation, an effect also observed when combined with TERB. These findings indicate that norepinephrine interaction with immune cell α- and β-ARs have opposing influences on lymphocyte proliferation, with α-ARs promoting and β-ARs opposing lymphocyte expansion during the effector phase of AA. Studies examining the effects of α-AR agonists on lymphocyte proliferation have reported increased, decreased, or no effect [reviewed in Ref. (48)]. Given the current evidence to date, lymphocytes express primarily β₂-ARs, while innate immune cells appear to express β₂-, α₁-, and α₂-ARs. Thus, effects of the α-AR antagonist, PHEN, on lymphocyte proliferation are likely mediated indirectly via α-AR-induced changes in functions of innate immune cells, such as macrophages, and/or by altering norepinephrine availability through presynaptic α₂-ARs. Given the multiple potential targets of α-AR ligand, the mixed results regarding α-AR-mediated effects on proliferation is not surprising. Discrepancies regarding the direction of the impact of α-ARs-targeting drugs on lymphocyte proliferation are likely due to experimental differences, such as the type of antigen challenge, in vivo or in vitro drug administration, strain differences, and timing of administration relative to immune challenge. Collectively, our findings support that in contrast to β₂-ARs, α-ARs suppress lymphocyte cell proliferation during the effector phase of AA, an effect that is blocked by α-AR antagonists during the effector phase. The extent to which the effects of TERB, PHEN or PT on proliferative responses contributes to disease reduction is unclear. However, inhibition of clonal expansion of arthritogenic T cells would be expected to reduce disease severity. Future studies will explore this possibility, as well as, the cell types targeted by these adrenergic treatments.

In contrast to TERB, PHEN treatment increases the production of IL-2 in spleen cells. Interestingly, PT also elevated IL-2 in DLN cells, but had no effect in other lymphoid compartments. The importance of these findings for reduced pathology is not clear. IL-2 is a pleiotropic cytokine. Besides its potent T-cell growth factor activity, IL-2 is essential for promoting the differentiation of Th1 (49) and Th2 cells (50). While IL-2 inhibits Th17 (51) and T follicular helper (Tfh) cell development (52), it does promote Th17 cell expansion once these cells develop (53). IL-2 is also involved in activation-induced cell death (AICD), which is important for homeostasis and the elimination of potentially harmful auto-reactive cells, at least in part by a Fas and FasL-dependent mechanism (54). IL-2 promotes antibody production and proliferation by B cells (55) and drives the development of CD4+FOXP3+ regulatory T cells (Treg cells), which have suppressor functions and mediate tolerance (56–58). Thus, IL-2 has broad essential biological actions, not only driving T cell proliferation and modulating effector cell differentiation, but also limiting potentially dangerous autoimmune reactions. Further research is required to understand the impact of α- and β-AR-mediated changes in IL-2 production on disease outcome in our disease model.

Regarding IL-2 and leukocyte proliferation responses, our findings are consistent with reports of increased SNS activity, decreased IL-2 production, and reduced proliferative responses in RA patients. Serum levels of IL-2 are reduced in patients with active disease, particularly in those with extra-articular manifestations (59, 60). Further, in RA patients mitogen-stimulated PBMCs produce less IL-2 and proliferate less robustly than PBMCs from controls (59). Low IL-2 and proliferative responses inversely correlate with disease activity and positively correlate with the proportion of CD4+ T cells (59), suggesting their importance in the pathophysiology of RA. Moreover, monocytes may play a role in regulating IL-2, as their depletion and partial reconstitution increases IL-2 production and proliferation in RA patients. Circulating cytokines that are indicators of general immune activation (including IL-2) increase prior to disease onset in RA patients (61), consistent with their importance in disease onset. Our data showing the ability of an α-AR antagonist or combined β₂-AR agonist and α-AR antagonist treatment to increase splenic and DLN IL-2, respectively, indicate the complexity of regulatory mechanisms for this cytokine by the SNS. IL-2 is at the “cross-roads” of both effector T-cell responses and tolerance [reviewed in Ref. (62)], and the development of lethal autoimmune disease in IL-2 knock-out mice (63). We have speculated that this particular cytokine may be the key to explaining opposite effects of adrenergic agonists on disease severity when administered prior to vs. after disease onset (12). If this hypothesis is true, then understanding of the mechanisms for these contradictory effects of adrenergic treatments relative to disease onset has implications for prevention and therapeutic treatment of RA.

In the present study, TERB treatment had no effect on IFN-γ production in spleen, MLN and PBMC, despite the well accepted role of norepinephrine via β₂-ARs to inhibit and enhance the production of the Th1 and Th2 cytokines, IFN-γ and IL-4, respectively (64, 65). The inability of TERB treatment to impact Th cell IFN-γ and IL-4 production in spleen, MLN, and PBMCs are consistent with findings by Heijink and coworkers (66) demonstrating that polarized Th1 and Th2 cells are less responsive to negative feedback by receptors coupled to the adenylate cyclase/cAMP pathway compared with freshly isolated T cells. Sanders and coworkers (67) also showed that in freshly isolated CD4+ T cells, a β₂-AR agonist could activate the cAMP pathway to increase IL-2 and IFN-γ, however, this response is attenuated in differentiated Th1 and Th2 cells. In these studies, specialized Th subset cells were stimulated by anti-CD3/anti-CD28 or in vitro differentiation of Th0 cells under Th1 or Th2 polarizing conditions. Similar to the differentiated Th cells, we also observed that treatment with β₂-AR agonists is no longer able to induce an increase in intracellular cAMP in splenocytes from arthritic rats 28 days post-adjuvant challenge (18). Further, lymphocyte β₂-ARs in spleen cells from arthritic rats were phosphorylated at a site known to induce receptor desensitization. Our findings are consistent with the loss of control over cytokine production by the β₂-AR-coupled cAMP pathway in Th effector cells after CFA challenge. These findings are also consistent with the differential effects on disease outcome in AA rats when a β-agonist is administered at different times across the time course of AA. Thus, differential effects of the agonist after CFA challenge and after disease onset may be due to targeting of differentiating Th0 cells versus fully differentiated Th cells.

Despite the known inhibitory effects of β₂-AR agonists on Th1 cell production of IFN-γ (64, 65), TERB treatment in arthritic rats elevated IFN-γ production in the DLNs. This finding suggests that β₂-AR signaling is not only uncoupled from its normal cAMP-PKA pathway, but also switches receptor coupling to an alternate second messenger. Recent findings from our lab indicate that in DLNs, β₂-ARs are phosphorylated at a site known to switch signaling from cAMP to mitogen-activated protein kinase (MAPK) pathways (18). In AA rats, DLNs drain a site of chronic inflammation, and thus, are exposed to high concentrations of inflammatory cytokines. This may explain, in part, the different β₂-AR responses in spleen and DLN cells. Consistent with this hypothesis, IL-1β is reported to cause a concentration- and time-dependent decrease in responses of airway smooth muscle cell and cardiac myocyte to a β-AR agonist that is mediated by uncoupling β-AR from Gs-induced activation of adenylyl cyclase (47, 68, 69). This response was accompanied by an increase in membrane Gi expression (69), a G-protein coupled with MAPK activation. The DLNs also receive a much higher antigen load than other secondary lymphoid organs after base of the tail CFA challenge. Additionally, the slow release of the M. butyricum cell wall components from the adjuvant may result in the persistent sub-optimal challenge in the DLNs. Interestingly, low persistent challenge with environmental adjuvant exposure has been linked to failed tolerance and the development of autoimmunity (70).

Neither TERB nor PHEN treatment altered IL-4 production in any of the tissues compared with VEH-treated arthritic rats. Our findings are consistent with studies showing that treatment with either norepinephrine or TERB induces IL-4 production from Th2 clones or Th2 effector cells derived from β₂-AR-stimulated naïve CD4+ T cells (67, 71, 72). In contrast, other studies have shown that elevations in cAMP were able to increase IL-4 production in Th2 cells (66). It is unlikely that β₂-ARs directly mediate increased IL-4 production by Th2 cells as the differentiation of Th0 to Th2 cells induces the loss of β₂-AR expression (67, 71, 72). Notably, PT treatment increased IL-4 production in the DLNs. Whether this is a direct or indirect result of TERB or PHEN, the result of an interaction between the two drugs on Th2 cell IL-4 production, or due to TERB- or PT-induced effects on other immune cell populations will require additional studies.

The most dramatic effects of PHEN treatment during the effector phase were the reductions of IFN-γ and IL-2 in the PBMCs. Since T cells do not normally express α-ARs, the effects of PHEN or PT on IL-2 or IFN-γ cytokine production could be indirect. PHEN binding to α₂-ARs present in presynaptic terminals is expected to increase availability of norepinephrine for interaction with T cell β₂-ARs. A change in circulating norepinephrine due to spillover from organs that are innervated and altered epinephrine release from the adrenal medulla could alter the circulating concentrations of norepinephrine and epinephrine. If this occurs, then circulating T cell β₂-AR stimulation could induce changes in T cell cytokines in this compartment. However, since TERB had no effect on PBMC IFN-γ release, the decreased IFN-γ production observed following PHEN or PT treatment is unlikely to be mediated indirectly through presynaptic α₂-ARs. Alternatively, these treatments could be targeting α-ARs expressed in monocytes to alter their production of cytokines that then induce changes in T cell cytokine production. This notion is consistent with a report by Heijnen and coworkers (73) showing functional α₁-ARs in peripheral blood leukocytes (presumably monocytes) of patients with polyarticular juvenile RA. Finally, the altered PBMC T cell cytokine production observed after PHEN or PT treatment could alter the T cell subtypes that enters the circulation, increasing subtypes that produce higher amounts of IFN-γ and IL-2. Mechanisms responsible for the effects of PHEN treatment on IFN-γ and IL-2 production in PBMCs during the effector phase of AA requires further investigation.

In this study, TERB or PHEN treatment did not alter production of the pro-inflammatory cytokine, TNF-α in spleen, lymph nodes, or PBMCs in arthritic rats after disease onset. This is in contrast to reports that β₂-AR agonists and α-AR antagonists inhibit macrophage TNF-α production, respectively [reviewed in Ref. (74–77)]. These findings suggest an impairment of signaling via both β₂- and α-ARs that occurs during the effector phase of AA. Interestingly, treatment with PT significantly reduced production of TNF-α in the spleen, MLN, and PBMCs. It is not clear why the combination drug treatment reduces TNF-α, in the absence of effects by the individual treatments, but the findings suggest interactions between β₂- and α-AR signaling. Collectively, these findings indicated that SNS signaling to immune cells via β₂- and α-ARs is impaired, or induces alternative second messengers in monocytes/macrophages in these organs in rats with established AA.

The greatest effect of TERB and/or PHEN on cytokine production was observed for splenic IL-10 production, where all three adrenergic treatments induced a significant increase in IL-10 production. This finding is consistent with the known effects of β₂-AR agonists and α-AR antagonists to increase IL-10 production (78, 79). The combination treatment increased IL-10 production to the same extent as treatment with each drug alone, suggesting that the effects of each of the components were neither additive nor synergistic. Thus, at least in the spleen, these receptors are capable of inducing production of second messengers, despite the lack of AR-induced responses for TNF-α production. In DLNs, IL-10 production increased significantly only after treatment with combined TERB and PHEN, although there was a trend for increased IL-10 with each individual adrenergic drug. We observed no effect for any of the adrenergic treatments in IL-10 production by PBMCs. In contrast, PHEN significantly reduced IL-10 levels in the MLNs, an effect that is inconsistent with reports that α-AR agonists suppress IL-10 production (78). Thus, the pattern of IL-10 production after adrenergic treatments was dependent upon the secondary lymphoid tissue examined.

Data supports that sympathetic dysfunctions significantly alter the balance between Th1-Th2 cell differentiation in RA (18, 80). RA is dominated by a Th1 response with selective accumulation of Th1 cells within the synovial compartment (81). RA and juvenile chronic arthritis co-exist with chronically elevated sympathetic activity [reviewed in Ref. (8)]. We and others have demonstrated that sympathetic innervation of the lymphoid organs in arthritic rodents (14, 15) and in arthritic joints of RA patients (17, 82) are lost and/or reorganized. Indeed, in this study, treatment with a β₂-AR agonist fails to shift T lymphocyte cytokine responses toward a Th2 profile in any of the secondary lymphoid organs of AA rats. However, in vivo treatment with the α-AR antagonist shifts the ex vivo cytokine profile from Th1 to a Th2 response, indicating that α-AR stimulation inhibits this shift. These findings are consistent with a recent report showing that catecholamines fail to shift T cell cytokines from a Th1 to a Th2 profile in PBMCs from RA patients, as seen in healthy donors (83). Thus, impaired sympathetic functioning is intimately linked with disease pathology.

Stimulating β₂-ARs and/or blocking α-ARs reduces the ratio of pro-to-anti-inflammatory cytokine production by PBMCs, spleen, and DLN cells. The reduced pro-to-anti-inflammatory cytokine ratios are consistent with the well-known effects of β₂-AR stimulation or α-AR blockade to inhibit TNF-α and promote IL-10 production in macrophages and dendritic cells (74–79). Spleen cells displayed the greatest drug-induced changes in anti-inflammatory profiles, largely due to greater IL-10 production. In contrast, TERB or PHEN treatment, tend to induce an anti-inflammatory secretory profile in PBMCs and DLN cells. However, dual drug treatment that augment β₂- and dampen α-AR activation-induced anti-inflammatory profiles. These changes were due to reduced TNF-α secretion in PBMCs and increased IL-10 in DLN cells. Therefore, TERB, PHEN, or PT-induced anti-inflammatory profiles may be responsible, at least in part, for improved disease outcomes in our study.

Interestingly, PHEN treatment induced a pro-inflammatory cytokine profile in MLN cells, an effect due to lower IL-10 in the absence of an effect on TNF-α production. Beta₂-AR stimulation inhibited TNF-α secretion, but was unable to significantly reduce the pro-to-anti-inflammatory ratio. Thus, blocking α-AR activation in the MLNs had an opposing effect on IL-10 production to that seen in the spleen. The reason for this difference may lie in the cellular source of IL-10 in these immune organs, but requires further investigation. Moreover, our data suggest that activation of either α- or β-ARs have anti-inflammatory effects in the MLNs. Why our drug treatments differentially affected inflammatory drive depending on the specific lymphoid tissue is unclear and warrants further study. Perhaps these differences reflect the unique functions of each lymphoid organ/tissue in response to the CFA challenge and the site of injection. The primary function of DLNs is to respond to the intradermal challenge, whereas the spleen participates in the systemic response. As part of the systemic response in AA, the spleen filters the blood, and is a major source of arthritogenic T cells that infiltrate affected joints. In contrast, the blood serves as a conduit for trafficking cells of the immune system. Finally, the MLNs are most responsive to antigens that travel in lymphatic vessels from the gut. Collectively, our findings indicate that the SNS maintains the ability to inhibit (via β₂-ARs) or promote (via α-ARs) inflammation in PBMCs, splenocytes, and DLN cells, but not in MLN cells.

Body weights are reduced 28 days after disease induction. This finding concurs with the well-documented AA-, CIA- and RA-induced cachexia (loss of body mass without altered diet or increased activity) due to inflammation that causes hypermetabolic protein and fat breakdown, skeletal muscle wasting, and to a lesser extent, lost white adipose tissue mass (12, 84–86). Moreover, treatment with PHEN, TERB, or PT significantly reduced disease-associated cachexia. The significance of these findings is underscored by cachexia occurring in ~2/3rds of all RA patients, and is a major contributor in greater morbidity and mortality in RA patients (84, 87, 88). Here, we did not monitor food intake, but other studies have shown that this factor is not contributory to disease-induced weight loss (89). However, inflammation indirectly reduces cellular mass by protein breakdown due to activation of the ubiquitin–proteasome pathway (90) and lipolysis (86). These effects are mediated at least in part by TNF-α (91). Adrenergic therapies in our study suppresses ex vivo TNF-α production and the pro- to anti-inflammatory ratios in a treatment- and immune organ-specific manner, consistent with a role for inflammation in rheumatoid cachexia.

In summary, it is clear that the critical functions of the SNS to regulate lymphocyte proliferation, Th1/Th2 cell differentiation, and inflammation are impaired/altered in AA rats, due to altered β₂- and α-AR functions. Selective α-AR blockade and/or β-AR activation in AA suppresses ex vivo proliferation, but is independent of sympathetic regulation of IL-2 production in AA. This is true in all immune organs examined, except PBMCs, and is consistent with the function of blood as a conduit for immune cell trafficking. Adrenergic treatments affect Th1/Th2 cytokines differently in every tissue examined in AA, perhaps reflecting the different immune functions of each organ. Most striking was the β₂-AR-mediated increase in IFN-γ in the DLNs and lack of a β₂-AR-mediated effect on IFN-γ in the spleen, particularly since arthritogenic T cells are generated at both sites (92, 93). Regardless of these differences, the overall outcome is a failure to drive Th2-type cytokine expression via β₂-AR stimulation, and thus, loss of a negative feedback mechanism for regulating cellular immunity. These findings are consistent with reports of disease-induced elevation in sympathetic tone (8), and high norepinephrine concentration-dependent β₂-AR down-regulation, desensitization, and/or shifts in G-protein coupling (18, 82, 94, 95). Indeed, we have previously discovered differences in phosphorylation of β₂-ARs in DLN and spleen cells collected from AA rats that may explain the IFN-γ findings in this study (18). Further, essential SNS regulation of innate immunity, via β₂- and α-ARs, is also impaired as demonstrated, with inability of β₂-AR stimulation or α-AR blockade to reduce TNF-α and increase IL-10 production in DLN cells. Stimulation of β₂-ARs similarly fails to reduce TNF-α production in the spleen cells.

Collectively, our findings indicate disrupted β₂-ARs, but not α-ARs signaling in AA, which consequently derails sympathetic regulation of lymphocyte expansion, Th cell differentiation, and inflammation required for immune system homeostasis. The clinical relevance of our findings is several-fold. First, they underscore the complexity of the disease’s relevant immune functions, their regulation by the SNS and the limitations of peripheral blood to decipher complex pathology. Secondly, distinct differences in SNS-immune interactions in spleen and lymph nodes likely reflect functional and microenvironmental differences of these immune organs, which currently are under appreciated and sources of discrepancies, inconsistencies, and confusion in the literature. Thus, relevant animal models for RA and novel approaches are required to elucidate both immune and neural-immune mechanisms in disease pathology. Finally, disease-modifying and suppressive inflammatory effects of targeting β₂- and α-ARs strongly support that combined AR-targeted therapies have great therapeutic potential to reduce joint destruction and inflammation and complications of RA, like cachexia. While promising, a greater understanding of the mechanisms by which these beneficial effects are realized is essential to evaluate therapeutic potential and guide clinical applications.

All authors made a substantial contribution to conceiving and/or designing of the research project, the acquiring, analyses, or interpretation of the data presented in this paper. All authors have contributed to writing of this paper and its intellectual content and critical evaluation of content. All authors have approved the final version of this paper, and are accountable for the accuracy and integrity of the research presented in this paper.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.