TNF-alpha downregulates murine hepatic growth hormone receptor expression by inhibiting Sp1 and Sp3 binding

Abstract

Children with chronic inflammatory diseases experience growth failure and wasting. This may be due to growth hormone resistance caused by cytokine-induced suppression of growth hormone receptor (GHR) gene expression. However, the factors governing inflammatory regulation of GHR are not known. We have reported that Sp1 and Sp3 regulate hepatic GHR expression. We hypothesized that TNF-alpha suppresses GHR expression by inhibiting Sp1/Sp3 transactivators. LPS administration significantly reduced murine hepatic GHR expression, as well as Sp1 and Sp3 binding to GHR promoter cis elements. TNF-alpha was integral to this response, as LPS did not affect hepatic Sp1/Sp3 binding or GHR expression in TNF receptor 1-deficient mice. TNF-alpha treatment of BNL CL.2 mouse liver cells reduced Sp1 and Sp3 binding to a GHR promoter cis element and downregulated activity of a GHR promoter-driven luciferase reporter. Combined mutations within adjacent Sp elements eliminated GHR promoter suppression by TNF-alpha without affecting overall nuclear levels of Sp1 or Sp3 proteins. These studies demonstrate that murine GHR transcription is downregulated by LPS, primarily via TNF-alpha-dependent signaling. Evidence suggests that inhibition of Sp transactivator binding is involved. Further investigation of these mechanisms may identify novel strategies for preventing inflammatory suppression of growth.

Figure 1

Endotoxin-induced TNF-α suppresses hepatic GHR gene expression. WT and TNF receptor 1–deficient mice were treated with LPS and total liver RNA was harvested after 12 hours. (a) Northern blots for total hepatic GHR and GHBP transcripts and (b) quantitative TaqMan RT-PCR for total and L2 hepatic GHR transcripts and a GAPDH internal control were performed. The relative amounts of each transcript in LPS-treated mice were calculated as a percentage of the corresponding value in control mice. The results are depicted as mean ± range (filled bar). The range is determined by evaluating the expression: 2^–ΔΔCT with ΔΔC_T + s and ΔΔC_T – s, where s = the SD of the ΔΔC_T value (18, 19). n = 3. ^AP < 0.01 (ANOVA) versus control. TNFR KO, TNF receptor 1 knockout mouse.

Figure 2

Sp1 and Sp3 bind the L2 GHR promoter via adjacent response elements. EMSA was performed using nuclear proteins from (a) adult mouse liver or (b and c) BNL CL.2 mouse liver cells and radiolabeled L2A or L2B elements. Supershift (a and b) or immunodepletion (c) with Sp1, Sp3, IgG, or STAT1 antibodies was performed. (c) EMSA for nuclear extracts that were immunodepleted using Sp1, Sp3, or control IgG antibodies before EMSA with the L2A element is shown on the left. Immunoblots confirming specific Sp1 or Sp3 depletion in these extracts are shown on the right. Ab: antibody used for immunodepletion; I: L2A or L2B complex containing both Sp1 and Sp3 proteins; II: L2A or L2B complex containing only Sp3; IIIA or IIIB: faster migrating complexes — L2A or L2B, respectively — which do not contain Sp1 or Sp3; HNF1: HNF1 consensus oligonucleotide.

Figure 3

Endotoxin-induced TNF-α downregulates L2 GHR promoter Sp transactivators. WT and TNF receptor 1–deficient mice were treated with LPS, and liver nuclear proteins were isolated after 12 hours. EMSA was performed using L2A, L2B, or HNF1 probes. (a) LPS-induced changes in binding to L2A or L2B complexes I and II in WT mice. (b) LPS-induced changes in binding to L2A or L2B complexes IIIA and IIIB in WT mice. (c) Effects of LPS treatment on binding to L2A or L2B in TNF receptor 1–deficient mice. (d) LPS-induced changes in binding to the HNF1 element in WT and TNF receptor 1–deficient mice. Changes in signal intensity were quantified by densitometry. The mean value for each group of three mice is shown. ^AP < 0.05. C, control; TNFR KO, TNF receptor 1 knockout mouse; L, LPS-treated mouse.

Figure 4

TNF-α downregulates GHR L2 promoter activity via the L2A and L2B Sp response elements. (a) WT and mutated GHR L2 promoter luciferase reporter constructs were transiently transfected in mouse liver BNL CL.2 cells and treated with TNF-α (10 ng/ml for 12 hours). Luciferase-specific activity in the cell homogenates was equalized for transfection efficiency monitored by cotransfection of a plasmid expressing β-gal (RLU/βGAL). Data are expressed as the mean ± SEM of three independent transfections performed in triplicate. ^AP < 0.05 versus control for each plasmid with TNF treatment. ^BP < 0.05 versus L2 control plasmid for basal activity. (b) EMSA was performed using mouse liver nuclear proteins and L2A, L2B, L2M1, and L2M3 oligonucleotides.

Figure 5

TNF-α suppresses transcription factor binding to the GHR L2A and L2B response elements. BNL CL.2 cells were treated with TNF-α (10 ng/ml for 4–24 hours), and nuclear proteins were isolated. EMSA was performed using oligonucleotides corresponding to the L2A and L2B response elements. Changes in signal intensity were quantified by densitometry and are shown. ^AP < 0.05.

Figure 6

TNF-α does not alter overall nuclear levels of the Sp1 or Sp3 proteins. BNL CL.2 cells were treated with TNF-α (10 ng/ml for 6–24 hours), and nuclear proteins were isolated. Immunoblots were performed using polyclonal Sp1 or Sp3 antibodies as described in Methods. A representative blot is shown; n = 3–5.