The growth hormone receptor interacts with transcriptional regulator HMGN1 upon GH-induced nuclear translocation

Abstract

Growth hormone (GH) actions are mediated through binding to its cell-surface receptor, the GH receptor (GHR), with consequent activation of downstream signalling. However, nuclear GHR localisation has also been observed and is associated with increased cancer cell proliferation. Here we investigated the functional implications of nuclear translocation of the GHR in the human endometrial cancer cell-line, RL95-2, and human mammary epithelial cell-line, MCF-10A. We found that following GH treatment, the GHR rapidly translocates to the nucleus, with maximal localisation at 5-10 min. Combined immunoprecipitation-mass spectrometry analysis of RL95-2 whole cell lysates identified 40 novel GHR binding partners, including the transcriptional regulator, HMGN1. Moreover, microarray analysis demonstrated that the gene targets of HMGN1 were differentially expressed following GH treatment, and co-immunoprecipitation showed that HMGN1 associates with the GHR in the nucleus. Therefore, our results suggest that GHR nuclear translocation might mediate GH actions via interaction with chromatin factors that then drive changes in specific downstream transcriptional programs.

Keywords: GHR; HMGN1; Mass spectrometry; Nuclear; Transcription factor.

Fig. 1

Characterisation of GH response in RL95-2 cells. a Semi-quantitative RT-PCR analysis of GHR and PRLR mRNA expression in RL95-2, MCF-10A, and the breast cancer cell line MCF7 (positive control). b GH treatment time-course in RL95-2 cells. Serum-starved RL95-2 cells were treated with 500 ng/ml recombinant human GH for 0, 5, 10, 15 and 30 min. c GH treatment concentration–response in RL95-2 cells. Serum-starved RL95-2 cells were treated with 0, 50, 100, 250 and 500 ng/ml GH for 15 min. Cell lysates were immunoblotted for phosphorylated STAT5 (pSTAT5) and total STAT. β-ACTIN was used as a loading control for all experiments

Fig. 2

Subcellular localisation of GHR in RL95-2 cells following GH treatment. a RL95-2 cells were grown on coverslips, serum-starved and treated with 500 ng/ml recombinant human GH for 0, 5, 10, 15 and 30 min. Cells were then fixed, permeabilised, blocked and immuno-stained with anti-GHR antibody and fluorescent secondary antibody. The slides were visualised using confocal laser scanning microscopy. Green (Alexa-Fluor 488) represents GHR staining (anti-GHR_IC antibody, sc-137185) and blue (DAPI) is a nuclear stain. Scale bar is 100 µm. b Quantification of cells with nuclear GHR localisation at different time points. Data are presented as mean ± SEM (n > 100 per timepoint). Groups with different letters are significantly different from each other (P < 0.001, One-way ANOVA). This experiment was repeated at least three times and a representative figure is shown

Fig. 3

a Volcano plot showing proteins identified by co-immunoprecipitation with GHR coupled with mass spectrometry that were differentially enriched following treatment with 500 ng/ml GH for 5 min, compared to control (log2FC). Blue dots represent proteins significantly depleted in treated samples compared to the control; red dots proteins either significantly enriched compared to the control following GH treatment or only present in the GH-treated group. b Volcano plot representation of differentially expressed genes between control and treatment with 500 ng/ml GH for 90 min (log2FC) obtained using a Clariom D microarray. Blue dots represent significantly downregulated genes (61 genes) and red dots upregulated genes (355 genes)

Fig. 4

a Venn diagram showing the intersection of known gene targets of HMGN1, SUMO1, and STAT5 obtained from the ChIP-atlas database with differentially expressed genes from microarray analysis. b Investigation of nuclear association of the GHR with HMGN1 by co-immunoprecipitation followed by western blotting. Serum-starved RL95-2 cells were treated with 500 ng/ml recombinant human GH for 0 and 5 min, lysed, and cytoplasmic and nuclear fractions isolated. Proteins were immunoprecipitated with the combination of extracellular and intracellular (ab89400 and sc-137185) anti-GHR antibodies or an HMGN1 antibody. Eluates were separated by SDS-PAGE and immunoblotted using antibodies to HMGN1 or GHR (ab89400) as indicated. HDAC1 was used as a nuclear control