Low- versus high-baseline epinephrine output shapes opposite innate cytokine profiles: presence of Lewis- and Fischer-like neurohormonal immune phenotypes in humans?

Abstract

Immunogenetic mechanisms operating within the immune system are known to influence cytokine profiles and disease susceptibility. Yet the role of the individual's neurohormonal background in these processes remains undefined. Hormonal imbalances are documented in immune-related diseases, but it is unclear whether this represents a secondary phenomenon or a primary "defect" related to specific neurohormonal immune phenotype(s). We report that in a large subpopulation of healthy humans the baseline epinephrine output (but not cortisol and sex steroid hormones) correlated inversely with proinflammatory and positively with anti-inflammatory cytokine production. Thus, low vs high epinephrine excretors had a 2- to 5-fold higher TNF-alpha and IL-12 production but 2-fold lower IL-10 production induced by LPS ex vivo. In alternative settings, we found low baseline levels and profoundly blunted stress-induced epinephrine responses but high TNF-alpha levels in Lewis vs Fischer inbred rats. Additionally, isoproterenol, a beta adrenoreceptor agonist suppressed LPS-induced TNF-alpha production, with more pronounced effect in Lewis than in Fischer rats. In human monocytes, epinephrine and the beta(2) adrenoreceptor agonist fenoterol potently inhibited LPS-induced TNF-alpha and IL-12, but stimulated IL-10 production. The order of potency for hormones able to inhibit IL-12 production ex vivo was: epinephrine > norepinephrine > or = 1,25-(OH)(2) vitamin D(3) > hydrocortisone. This indicates that baseline epinephrine conditions cytokine responsiveness and through this mechanism intrinsic hypo- or hyperactive adrenal medullas in some individuals may shape opposite cytokine profiles. Since Lewis and Fischer rats have opposite susceptibility to experimental immunological diseases, this suggests that the parallel human phenotypes could be linked to differing responsiveness and susceptibility to infections and immune/inflammatory-related conditions.

FIGURE 1.

Reproducibility of human IL-12 production interindividual variations. LPS-induced IL-12p70 production in whole blood cultures ex vivo from seven healthy subjects, as determined by ELISA and described in Materials and Methods. Experiments with whole blood cultures from different individuals were repeated two to seven times within several weeks, months, and, in some cases, within 2–3 years. To avoid circadian variations of cytokine production, experiments were performed between 2 and 3 p.m. Raw data for the cytokine production in pg/ml are depicted for each subject.

FIGURE 2.

Correlations among IL-12, TNF-α, and IL-10 production ex vivo in 107 healthy human subjects. A, Plot of IL-12 vs TNF-α. B, Plot of IL-12 vs IL-10. C, Ex vivo IL-12 production in low TNF-α (≤3,423 pg/10⁶ monocytes) vs high TNF-α producers (≥11,063 pg/10⁶ monocytes). D, Ex vivo IL-12 production in low IL-10 (≤236 pg/10⁶ monocytes) vs high IL-10 producers (≥954 pg/10⁶ monocytes). Low vs high TNF-α and IL-10 producers were estimated by subtracting or adding 1 SD from the means of the population estimates of TNF-α and IL-10 production values. Data are expressed as means ± SE. *** and ** designate significance at p < 0.001 and 0.01, respectively. Cytokine production was corrected for monocyte number and expressed as pg/10⁶ monocytes. The means of the population estimates for the ex vivo IL-12, TNF-α, and IL-10 production were 113, 7243, and 595 pg/10⁶ monocytes, respectively.

FIGURE 3.

Correlations between cytokines ex vivo and baseline EPI output in vivo in 107 healthy human subjects. A, Plot of IL-12 vs 24-h EPI excretion. B, Plot of TNF-α vs 24-h EPI excretion. C, Plot of IL-10 vs 24-h EPI excretion. D, Dose-dependent inhibitory effect of EPI and fenoterol on human IL-12 production in whole blood cultures. Ex vivo IL-12 (E), TNF-α (F), IL-10 (G), and IL-12:IL-10 ratio (H) in low (≤1.94 μg/24 h) vs high EPI excretors (≥7.19 μg/24 h). Low vs high EPI excretors were estimated by subtracting and adding 1 SD from the means of the population estimates of 24-h urinary EPI excretion. Data are expressed as means ± SE. ***, **, and * designate significance at p < 0.001, 0.0,1 and 0.05, respectively. Cytokine production was corrected for monocyte number and expressed as pg/10⁶ monocytes. The means of the urinary EPI excretion (μg/24 h ± SD) in low and high EPI excretors were 1.49 ± 0.34 and 10.13 ± 2.21, respectively. The means ± SD of the monocyte percentages and absolute monocyte count in low vs high EPI excretors were 7.39 ± 1.85% and 0.41 ± 0.14 × 10³/μl vs 7.44 ± 1.56% and 0.47 ± 0.12 × 10³/μl, respectively. The means of the population estimates for monocyte percentages and absolute monocyte count were 7.73% and 0.43 × 10³/μl. The means of the population estimates for 24-h urinary EPI excretion, the ex vivo IL-12, TNF-α, and IL-10 production and IL-12: IL-10 ratio were 4.57 μg/24 h, 113, 7243, and 595 pg/10⁶ monocytes and 0.37, respectively.

FIGURE 4.

Cytokine production, CA levels and responses in LEW vs F344 inbred rats. A, EPI baseline plasma levels. B, NE baseline plasma levels. C, LPS-induced TNF-α plasma levels. D, Splenic total RNA as assessed by Northern blot probed with TNF-α. E, EPI plasma levels after immobilization. F, NE plasma levels after immobilization. Data are expressed as means ± SE, five to seven animals per group. ***, **, and * designate significance at p < 0.001, 0.01, and 0.05, respectively. F, F344 rats; L, LEW rats.

FIGURE 5.

Effect of isoproterenol and propranolol on LPS-induced TNF-α production in LEW and F344 inbred rats. Data are expressed as means and SE, five to seven animals per group. *** and * designate significance at p < 0.001 and 0.05, respectively. ISO, Isoproterenol; PRO, propranolol.

FIGURE 6.

A schematic interpretation of the role of baseline EPI (adrenaline) in shaping innate cytokine responsiveness and cytokine profiles. A, In humans, a relative low activity of the adrenal medulla in some individuals provides low tonic inhibitory input on proinflammatory cytokine production, exerted by endogenous basal EPI and thus conditions peripheral monocytes to produce more IL-12 and TNF-α but less IL-10 upon antigenic stimulation than monocytes in individuals with relatively high adrenomedullary activity. B, Genetically determined hypoactive and hyperactive hypothalamic-pituitary-adrenal (HPA) axis and sympathetic nervous system (SNS) result in low vs high tonic inhibitory input on proinflammatory cytokine production exerted by endogenous basal GCs and CAs and thus determine to a great extend high vs low proinflammatory/Th1 responses in LEW and F344 rats, respectively. Solid lines represent stimulation, while dashed lines represent inhibition.