A novel effect of growth hormone on macrophage modulates macrophage-dependent adipocyte differentiation

Abstract

The GH receptor (GHR) is expressed on macrophages. However, the precise role of GH in regulation of macrophage function is unclear. We hypothesized that soluble factors including cytokines produced by macrophages in a GH-dependent manner regulate adipogenesis. We confirmed expression and functional integrity of the GHR in the J774A.1 macrophage cells. Conditioned medium (CM) from macrophages inhibited adipogenesis in a 3T3-L1 adipogenesis assay. CM from GH-treated macrophages decreased the inhibitory effect of CM from macrophages on adipogenesis. This effect on preadipocyte differentiation was active only during the first (early) phase of adipocyte differentiation. CM from stromal vascular compartment macrophages of mice with macrophage-specific deletion of the GHR exhibited more inhibitory effect on 3T3-L1 preadipocyte differentiation compared with CM from stromal vascular compartment macrophages of control mice, indicating that intact GH action in primary macrophages also increases preadipocyte differentiation. GH did not increase IGF-1 expression in macrophages. PCR array analysis identified IL-1beta as a candidate cytokine whose expression was altered by GH in macrophages. Levels of IL-1beta mRNA and protein were significantly decreased in GH-treated J774A.1 macrophages. Nuclear factor-kappaB stimulates IL-1beta gene expression, and GH induced a significant decrease in the levels of phosphorylated nuclear factor-kappaB in macrophages. IL-1beta is a known inhibitor of adipogenesis, and these results support GH-dependent down-regulation of macrophage IL-1beta expression as one mechanism for the observed increase in adipogenesis with CM from GH-treated macrophages. We conclude that GH decreases secretion of IL-1beta by the macrophage and thus in a paracrine manner increases adipocyte differentiation. These results provide a novel mechanism for GH's actions in the control of adipogenesis.

Figure 1

J77A4.1 cells express functional GHRs. A, Expression of GHR in J774A.1 macrophages. RNA was extracted from liver or J774A.1 cells and steady-state abundance of the major (L2) transcript of the GHR gene measured by TaqMan RT-PCR. Expression of the housekeeping gene GAPDH was used as an internal control to normalize results. The relative amounts of the transcript in J774A.1 cells are depicted relative to the expression in liver. The results (n = 3) are depicted as mean ± sem. B, Equal amounts of protein from F442A cells, GHR^−/− liver or J774A.1 cells were size-fractionated by 4–15% SDS-PAGE and immunoblotted with anti-GHR AL47 antibody. Specific signals were visualized using the enhanced chemiluminescence (ECL) system. The position of the specific GHR (∼115 kDa) and nonspecific (NS; ∼82 kDa) bands are indicated. Bottom panel, Overexposure of the immunoblot to demonstrate absence (GHR^−/− liver) or presence (J774A.1 cells) of GHR expression. C, GH stimulates canonical GHR signaling pathways in macrophages. J774A.1 macrophages were starved overnight and then exposed to ovine GH (500 ng/ml) for the indicated time periods. Whole-cell protein extracts were prepared and equal amounts of protein size fractionated by SDS-PAGE. Immunoblotting was performed with anti-phospho-JAK2, -JAK2, -phospho-ERK antibody, or -total ERK antibody. Specific signals were visualized using the ECL system. The identity of the protein is indicated.

Figure 2

Stimulation of macrophages with GH decreases the inhibitory activity of conditioned medium (CM) from macrophages on differentiation of 3T3-LI preadipocytes. A and B, 3T3-L1 preadipocytes were induced to differentiate in the presence of 0–100% of CM-Mac or GH-CM-Mac. Eight days after differentiation, the cultures were stained with Oil Red O (A) and the fat content quantified by eluting Oil Red O with isopropanol and measuring the OD at 492 nm (B; mean ± sem, n = 4). *, P < 0.05. C, PPARγ, aP2, and adiponectin mRNA expression. Total RNA was isolated from 3T3-L1 adipocytes (8 d after differentiation) grown in regular CM (control), CM-Mac, or GH-CM-Mac. PPARγ, aP2, and adiponectin mRNA abundance was measured by Syber Green RT-PCR analysis. Expression of the housekeeping gene GAPDH was used as an internal control to normalize results. The results (n = 3–5) are depicted as mean ± sem. The steady-state abundance of the transcript is depicted relative to the abundance in cells grown in regular culture medium. NS, Not significant.

Figure 3

The effect of conditioned medium (CM) from GH-treated macrophages on the differentiation of 3T3-L1 preadipocytes is restricted to the induction phase of adipogenesis. 3T3-L1 cells were incubated with CM (mixed 1:1 with culture medium) from either macrophages (CM) or GH-CM-Mac for the first 2 d (A) or last 6 d (B) after addition of the differentiation cocktail mixture. Differentiation of adipocytes was monitored by staining with Oil Red O staining 8 d after differentiation. C, Control.

Figure 4

The absence of GHR diminishes the salutary effect of conditioned medium from SVC macrophages on the differentiation of 3T3-L1 preadipocytes. A, Expression of GHR in GHRMacD [Ghr^{Δ/fl; LysM-cre (pos)}] and control [Ghr^{Δ/fl; LysM-cre (neg)}] mice. RNA was extracted from liver or splenocytes isolated from GHRMacD and control mice and steady-state abundance of GHR mRNA measured by TaqMan RT-PCR. Expression of the housekeeping gene GAPDH was used as an internal control to normalize results. The results (n = 3) are depicted as mean ± sem. The relative amount of the transcript is depicted relative to the expression in the control mice. B, The effect of conditioned medium from SVC macrophages on the differentiation of 3T3-L1 preadipocytes. 3T3-L1 cells were incubated with normal culture medium or conditioned medium from either bone marrow-derived macrophages (CM-Mac-BM) or SVC macrophages (CM-Mac-SVC) isolated from either GHRMacD [Ghr^{Δ/fl; LysM-cre (pos)}] or control [Ghr^{Δ/fl; LysM-cre (neg)}] mice. Differentiation of adipocytes was monitored by Oil Red O staining 8 d after differentiation. The results are representative of results obtained with two separate isolations of SVC macrophages from the respective mouse models. NS, Not significant.

Figure 5

GH’s effects on macrophage are IGF-I independent. Total RNA was isolated from either J774A.1 or primary murine macrophages (from adipose tissue SVC) exposed to GH (500 ng/ml for 24 h). IGF-I mRNA abundance was measured by Syber Green RT-PCR analysis. Expression of the housekeeping gene GAPDH was used as an internal control to normalize results. The results (n = 3–5) are depicted as mean ± sem. The amounts of the transcript are depicted relative to the expression in cells not exposed to GH. NS, Not significant.

Figure 6

GH-dependent decrease in IL-1β expression and secretion in macrophages. A, J774A.1 macrophages were stimulated with either vehicle or GH (500 ng/ml) for 24 h and total RNA extracted and subjected to analysis by SuperArray. A three-dimensional representation of the average changes (positive and negative) for two independent experiments is depicted. The bar representing IL-1β is indicated by an arrow. B, Total RNA was isolated from J774A.1 stimulated with either vehicle or GH (500 ng/ml) for 24 h, and IL-1β mRNA abundance was measured by Syber Green RT-PCR analysis. Expression of the housekeeping gene GAPDH was used as an internal control to normalize results. The results (n = 3) are depicted as mean ± sem. The amounts of the transcript are depicted relative to the expression in cells not exposed to GH. C, GH-dependent decrease in IL-1β expression in J774A.1 cells. J774A.1 cells were stimulated with either vehicle or GH (500 ng/ml) for 24 h, and the whole-cell lysates were size fractionated by SDS-PAGE. Immunoblotting was performed with anti-IL-1β or tubulin antibody. Specific signals were visualized using the enhanced chemiluminescence (ECL) system. The identity of the protein is indicated. D, IL-1β can inhibit adipogenesis at low concentrations. Adipogenesis assay were performed in 3T3-L1 cells treated with IL-1β at 0, 1, and 5 pg/ml concentration. Differentiation of preadipocytes was determined by Oil Red O staining 8 d after differentiation.

Figure 7

GH attenuates phosphorylation of NF-κB. Top panel, J774A.1 macrophages were starved overnight and then exposed to ovine GH (500 ng/ml) for the indicated time periods. Whole-cell protein extracts were prepared and equal amounts of protein size fractionated by SDS-PAGE. Immunoblotting was performed with anti-phospho-p65 NF-κB (Ser 536) and total NF-κB antibodies. To verify equivalence of sample loading, the abundance of tubulin was also measured by Western blot analysis. Specific signals were visualized using the enhanced chemiluminescence (ECL) system. The identity and molecular weights of the proteins are indicated. Bottom panel, Densitometric measurements (mean ± sem; n = 4) are depicted. *, P < 0.001 compared with 0 h value, which was assigned a value of 1.

Figure 8

Model for direct and indirect role of GH in adipogenesis. GH can act directly on the preadipocyte to influence adipogenesis. The studies in the current report also indicate a novel indirect mode of GH action on the preadipocyte wherein GH acts on the adipose tissue macrophage to decrease IL-1β secretion by the macrophage. IL-1β is a known inhibitor of adipogenesis, and thus, GH’s action on the macrophage serves to enhance adipogenesis.