Myeloid adrenergic signaling via CaMKII forms a feedforward loop of catecholamine biosynthesis

Abstract

Type 2 immune response has been shown to facilitate cold-induced thermogenesis and browning of white fat. However, whether alternatively activated macrophages produce catecholamine and substantially promote adaptive thermogenesis in adipose tissue remains controversial. Here, we show that tyrosine hydroxylase (TyrH), a rate-limiting enzyme of catecholamine biosynthesis, was expressed and phosphorylated in adipose-resident macrophages. In addition, the plasma level of adrenaline was increased by cold stress in mice, and treatment of macrophages with adrenaline stimulated phosphorylation of Ca2+/calmodulin-dependent protein kinase II (CaMKII) and TyrH. Genetic and pharmacological inhibition of CaMKII or PKA signaling diminished adrenaline-induced phosphorylation of TyrH in primary macrophages. Consistently, overexpression of constitutively active CaMKII upregulated basal TyrH phosphorylation, while suppressing the stimulatory effect of adrenaline on TyrH in macrophages. Myeloid-specific disruption of CaMKIIγ suppressed both the cold-induced production of norepinephrine and adipose UCP1 expression in vivo and the stimulatory effect of adrenaline on macrophage-dependent activation of brown adipocytes in vitro. Lack of CaMKII signaling attenuated catecholamine production mediated by cytokines IL-4 and IL-13, key inducers of type 2 immune response in primary macrophages. Taken together, these results suggest a feedforward mechanism of adrenaline in adipose-resident macrophages, and that myeloid CaMKII signaling plays an important role in catecholamine production and subsequent beige fat activation.

Keywords: CaMKII; UCP1; adrenaline; catecholamine; tyrosine hydroxylase.

Figure 1

Phosphorylation of TyrH is enriched in macrophages. (A) Surgical denervation markedly suppressed protein levels of TyrH in inguinal fat. Three 3-month-old male C57BL/6 mice were surgically denervated in inguinal fat and recovered for 2 weeks, and then euthanized after cold exposure for 1 day for western blot analysis of inguinal fat. S, sham; D, denervation. (B) The protein level of TyrH was low, while its phosphorylation was enriched in peritoneal macrophages compared to hypothalamus, brain stem, liver, heart, and adipose tissue of 3-month-old male C57BL/6 mice. Hypo, hypothalamus; BS, brain stem; BAT, brown adipose tissue; iWAT, inguinal white adipose tissue; eWAT, epididymal white adipose tissue; P-Mϕ, peritoneal macrophage; Non, non-specific band. (C) Cold stress stimulated the phosphorylation of TyrH in brown adipose tissue. TyrH in brown fat from mice exposed with or without cold was immunoprecipitated using anti-TyrH. For each immunoprecipitation reaction, 100 μg protein was used. Normal IgG was used as a negative control. Lysate from neuroblastoma N1E115, a catecholamine-producing clone, was used as a positive control for TyrH. (D) TyrH in peritoneal macrophages was immunoprecipitated for western blot analysis of TyrH protein and phosphorylation. Normal IgG was used as a negative control. Lys, lysate. (E) Immunofluorescence study of co-localization of macrophage marker F4/80 with P-TyrH in inguinal fat of mice under room temperature or cold stress condition. Scale bar, 50 μm. The data in E are the Representative images from at least three mice within each group. The data in B, C, and D are the representative from at least three individual experiments with similar results.

Figure 2

Treatment of adrenaline stimulates the phosphorylation of TyrH and CaMKII in macrophages. (A) Circulating level of adrenaline was elevated by 10 h cold stress (6°C) in the 3-month-old male mice. ELISA analysis was used to measure the level of adrenaline. (B) Adrenaline treatment at 20 nM and 100 nM induced UCP1 expression and PKA activation in differentiated brown adipocytes. CL316,243 was used as the positive control. (C) Analysis and quantification of UCP1 western blot in B. (D−G) Treatment of adrenaline stimulated the phosphorylation of TyrH at Ser⁴⁰ and CaMKII Thr²⁸⁶ in a dose (D) and time (E)-dependent manner in RAW264.7 macrophages. The dose (F) and time (G) course data for phosphorylation of TyrH and CaMKII in D and E were quantified using ScnImage software and normalized by the protein level of TyrH and CaMKII. * and ^# indicate the significance for P-CaMKII/CaMKII and P-TyrH/TyrH, respectively. (H) False color images showing the fluorescence ratio increased over time following exposure to adrenaline. (I) Averaged population data showing the time course of adrenaline-stimulated intracellular Ca²⁺ increase measured in RAW264.7 macrophages (mean ± SEM, n = 10 cells). Data were expressed as both a background-corrected fluorescence ratio (left y-axis, red for 340 nm and blue for 380 nm) and the estimated Ca²⁺ (right y-axis). The data in B, D, E, and H are the representative from at least three individual experiments with similar results. The data in A, C, F, G, and I are presented with mean ± SEM. *^,#P < 0.05, **P < 0.01, ***P < 0.005 compared with control.

Figure 3

PKA signaling stimulates the phosphorylation of TyrH in CaMKII-dependent manner in macrophages. (A and B) Short-time treatment of dbcAMP stimulated the phosphorylation of TyrH and CaMKII in a time (A) and dose (B)-dependent manner in RAW264.7 macrophages. (C) Treatment of 100 μM dbcAMP for 22 h induced the expression of TyrH as well as the phosphorylation of TyrH and CaMKII in RAW264.7 macrophages. (D) False color images showing the fluorescence ratio increases over time following exposure to dbcAMP in RAW264.7 macrophages. (E) Averaged population data showing the time course of dbcAMP-stimulated accumulation of intracellular Ca²⁺ measured in macrophages (mean ± SEM, n = 10 cells). Data were expressed as both a background-corrected fluorescence ratio (left y-axis, red for 340 nm and blue for 380 nm) and the estimated Ca²⁺ (right y-axis). (F) Inhibiting CaMKII by 10 μM KN93 blocked dbcAMP-stimulated phosphorylation of TyrH in primary peritoneal macrophages. (G) Western blot analysis and quantification of P-TyrH in F. (H) Inhibiting PKA by 10 μM H89 markedly suppressed dbcAMP-stimulated phosphorylation of CaMKII and TyrH in primary peritoneal macrophages. (I) Western blot analysis and quantification of P-TyrH in H. (J) Flow cytometry analysis of primary macrophages isolated from adipose tissue using magnetic beads. F4/80 positive cells were considered as macrophages, and stromal vascular fraction (SVF) was used to indicate the purity of macrophage in SVF before isolation. The staining of isolated macrophages without primary antibody was used as a negative control. (K) dbcAMP-stimulated phosphorylation of TyrH was suppressed by inhibiting CaMKII with treatment of 10 μM KN93 or inhibiting PKA with treatment of 10 μM H89 in primary adipose-resident macrophages. The data in A−D, F, H, J, and K are the representative from at least three individual experiments with similar results. The data in E, G, and I are presented with mean ± SEM. *P < 0.05, **P < 0.01.

Figure 4

CaMKII is required for PKA signaling-induced norepinephrine secretion in primary macrophages. Adipose-resident macrophages were treated with KN93 for 1 h followed by the treatment of dbcAMP for 4 h. The cells were then washed and cultured in 1 ml fresh serum-free medium for 12 h, and the media was collected to treat the differentiated brown adipocytes. (A) dbcAMP-induced norepinephrine secretion from adipose tissue macrophages were suppressed by CaMKII inhibitor KN93. Norepinephrine level was determined by ELISA. (B) CaMKII inhibition reversed the inducing effect of dbcAMP-treated macrophage media on UCP1 and C/EBPβ expression in brown adipocytes. (C) Western blot analysis and quantification of UCP1 and C/EBPβ in B. (D) CaMKIIγ deficiency suppressed dbcAMP-induced norepinephrine secretion from primary macrophages. (E) CaMKIIγ deficiency diminished the inducing effect of dbcAMP-treated macrophage media on phosphorylation of PKA and expression of UCP1 and C/EBPβ in brown adipocytes. (F) Western blot analysis and quantification of UCP1 and C/EBPβ in E. The data in A, B, D, and E are the representative from at least three individual experiments with similar results. The data in A, C, E, and F are presented with mean ± SEM. *P < 0.05, **P < 0.01.

Figure 5

CaMKII is required for adrenaline-induced norepinephrine secretion in macrophages. (A) Inhibiting CaMKII by treatment of 10 μM KN93 suppressed adrenaline-stimulated phosphorylation of TyrH in RAW264.7 macrophages. (B) Western blot analysis and quantification of P-TyrH in A. (C) Overexpression of CA-CaMKII elevated basal P-TyrH and diminished the stimulatory effect of adrenaline on P-TyrH in RAW264.7 macrophages. (D) Western blot analysis and quantification of P-TyrH in C. (E) Myeloid overexpression of CA-CaMKII induced UCP1 and C/EBPβ expression and diminished the inducing effect of adrenaline in co-cultured brown adipocytes. Adipose-resident macrophages were infected with adenoviruses encoding DN-CaMKII and CA-CaMKII for 24 h, and then changed into the fresh medium and treated with KN93 for 1 h followed by the treatment of dbcAMP for 4 h. The cells were then cultured in 1 ml fresh serum-free medium for 12 h, and the media was collected to treat the differentiated brown adipocytes. (F) Western blot analysis and quantification of UCP1 in E. (G) CaMKIIγ deficiency diminished adrenaline-stimulated phosphorylation of TyrH in peritoneal macrophages. The primary macrophages were isolated from CaMKIIγ KO and wild-type (WT) mice. (H) Western blot analysis and quantification of P-TyrH in G. The data in A, C, E, and G are the representative from at least three individual experiments with similar results. The data in B, D, F, and H are presented with mean ± SEM. *P < 0.05, **P < 0.01.

Figure 6

CaMKII plays an important role in IL-4/IL-13-induced catecholamine production in primary macrophages. The primary macrophages were cultured in serum-free medium containing 2% BSA for 4 h and then treated with or without KN93 for 1 h followed by the treatment of 10 ng/ml IL-4 and 10 ng/ml IL-13 for 22 h. (A) Phosphorylation of TyrH was stimulated by the treatment of adrenaline but not the treatment of IL-4 and IL-13 for 30 min in primary macrophages. (B) Expression and phosphorylation of TyrH were induced by 22 h treatment of IL-4/IL-13, and CaMKII inhibitor KN93 suppressed IL-4/IL-13-induced phosphorylation of TyrH in peritoneal macrophages. (C) Western blot analysis and quantification of P-TyrH in B. (D) CaMKIIγ deficiency diminished IL-4/IL-13-induced phosphorylation but not expression of TyrH in peritoneal macrophages. The primary macrophages were isolated from CaMKIIγ KO and WT mice. (E) Western blot analysis and quantification of P-TyrH and TyrH in D. (F) CaMKIIγ deficiency attenuated IL-4/IL-13-induced norepinephrine secretion from primary macrophages. The data in A, B, D, and F are the representative from at least three individual experiments with similar results. The data in C, E, and F are presented with mean ± SEM. *P < 0.05, **P < 0.01.

Figure 7

CaMKII deficiency suppressed basal activity of TyrH and UCP1 expression in inguinal not brown fat. (A) Cold exposure stimulated the phosphorylation of CaMKII in inguinal fat. The 3-month-old male C57BL/6 mice were housed in the metabolic phenotyping system with a temperature-controllable chamber and exposed to cold stress (6°C) or room temperature (22°C) condition for 48 h. (B) The ratio of phosphorylation to protein level of CaMKII in A was quantified. (C) Immunofluorescence study of co-localization of macrophage marker F4/80 with P-CaMKII in inguinal fat of mice under room temperature or cold stress condition. Scale bar, 50 μm. (D−J) The 6-week-old C57BL/6 mice were reconstituted with bone marrow cells from CaMKIIγ KO (BMT-KO) and WT mice (BMT-WT). Six weeks post bone marrow transplantation, indirect calorimetry and cold exposure were performed followed by euthanasia and fat tissue collection. PCR analysis was performed for genomic DNA from bone marrow-derived macrophages (BMDM) (D), bone marrow cells (BMC) and adipose tissue (E). The amplified band of smaller size (250 bp, top panel) indicates CaMKIIγ KO alleles, whereas the PCR product of larger size (500 bp, bottom panel) denotes the WT band. (F) CaMKIIγ-deficient bone marrow chimeras (BMT-KO) displayed downregulated phosphorylation of TyrH and expression of UCP1 in inguinal fat compared to control mice (BMT-WT) under cold stress condition. (G) Western blot analysis and quantification of P-TyrH and UCP1 in F. (H) The levels of P-TyrH and UCP1 in brown fat were similar between BMT-KO and BMT-WT mice under cold stress condition. (I) Western blot analysis and quantification of P-TyrH and UCP1 in H. (J) mRNA levels of UCP1 but not TyrH, IL-4 and IL-13 were significantly suppressed by myeloid-deficiency of CaMKIIγ in inguinal fat. All data are presented with mean ± SEM. *P < 0.05, **P < 0.01.