Signal transduction pathways activated by insulin-like peptide 5 at the relaxin family peptide RXFP4 receptor

Abstract

Background and purpose: Insulin-like peptide 5 (INSL5) is a two-chain, three-disulfide-bonded peptide of the insulin/relaxin superfamily, uniquely expressed in enteroendocrine L-cells of the colon. It is the cognate ligand of relaxin family peptide RXFP4 receptor that is mainly expressed in the colorectum and enteric nervous system. This study identifies new signalling pathways activated by INSL5 acting on RXFP4 receptors.

Experimental approach: INSL5/RXFP4 receptor signalling was investigated using AlphaScreen® proximity assays. Recruitment of Gα_i/o proteins by RXFP4 receptors was determined by rescue of Pertussis toxin (PTX)-inhibited cAMP and ERK1/2 responses following transient transfection of PTX-insensitive Gα_i/o C351I mutants. Cell proliferation was studied with bromodeoxyuridine. RXFP4 receptor interactions with β-arrestins, GPCR kinase 2 (GRK2), KRas and Rab5a was assessed with real-time BRET. Gene expression was investigated using real-time quantitative PCR. Insulin release was measured using HTRF and intracellular Ca²⁺ flux monitored in a Flexstation® using Fluo-4-AM.

Key results: INSL5 inhibited forskolin-stimulated cAMP accumulation and increased phosphorylation of ERK1/2, p38MAPK, Akt Ser⁴⁷³ , Akt Thr³⁰⁸ and S6 ribosomal protein. cAMP and ERK1/2 responses were abolished by PTX and rescued by mGα_oA , mGα_oB and mGα_i2 and to a lesser extent mGα_i1 and mGα_i3 . RXFP4 receptors interacted with GRK2 and β-arrestins, moved towards Rab5a and away from KRas, indicating internalisation following receptor activation. INSL5 inhibited glucose-stimulated insulin secretion and Ca²⁺ mobilisation in MIN6 insulinoma cells and forskolin-stimulated cAMP accumulation in NCI-H716 enteroendocrine cells.

Conclusions and implications: Knowledge of signalling pathways activated by INSL5 at RXFP4 receptors is essential for understanding the biological roles of this novel gut hormone.

Linked articles: This article is part of a themed section on Recent Progress in the Understanding of Relaxin Family Peptides and their Receptors. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v174.10/issuetoc.

Figure 1

Concentration–response relationships for activation of ERK1/2, Akt, p38MAPK and S6RP and inhibition of cAMP production by hINSL5 and mINSL5 in CHO‐RXFP4 cells. In (A) phosphorylation of ERK1/2 (5 min; n = 6), (B) Akt Ser⁴⁷³ (5 min; n = 6), (C) Akt Thr³⁰⁸ (5 min; n = 6), (D) p38MAPK (15 min; n = 5) and (E) S6RP (30 min; n = 5), with in (F) inhibition of forskolin‐stimulated cAMP accumulation [30 min; n = 5 (mINSL5), n = 7 (hINSL5)]. CHO‐RXFP4 cells were treated with increasing concentrations of hINSL5 or mINSL5 (10⁻¹¹ to 10^–6.5 M). Results are normalized to the maximum response to hINSL5 in protein phosphorylation assays or % of response to 3 μM forskolin in cAMP accumulation assay. Data points represent mean ± SEM of independent experiments. Veh, vehicle.

Figure 2

Effects of pathway inhibitors on signalling responses to mINSL5 in CHO‐RXFP4 cells. CHO‐RXFP4 cells were pretreated with the Gα_i/o inhibitor (PTX, 100 ng•mL⁻¹), MEK inhibitor (U0126, 10 μM), Src family tyrosine kinase inhibitor (PP2, 10 μM), PI3K inhibitor (LY294002, 10 μM) or the mTOR complex inhibitors (KU0063794, 1 μM; rapamycin, 100 nM) followed by INSL5 (100 nM). Pathways examined were (A) p‐ERK1/2, (B) Akt p‐Ser⁴⁷³ , (C) Akt p‐Thr³⁰⁸ and (D) p‐S6RP. Cells were treated with inhibitors for 30 min before and then during mINSL5 stimulation, except for the PTX treatment where cells were exposed for 18 h. Results are quantified as fold change in fluorescence over that of the vehicle treatment in the control group. Bars represent mean ± SEM of experiments [n = 5 (PTX, LY294002, KU0063794, rapamycin), n = 6 (PP2, U0126), n = 10 (Control)]. *P < 0.05; significantly different from the mINSL5 response in the control group; repeated‐measures two‐way ANOVA followed by Dunnett's multiple comparisons test. ns, non‐significant.

Figure 3

Determination of the Gα_i/o isoforms utilise by RXFP4 receptors using PTX‐resistant Gα mutants. cAMP inhibition and ERK1/2 activation following stimulation by mINSL5 in CHO‐RXFP4 cells. RXFP4 receptors inhibit forskolin‐stimulated cAMP accumulation, indicating an interaction with Gα_i/o proteins. CHO‐RXFP4 cells were transiently transfected with pcDNA3 (mock transfection) in (A, G) or G_i/o subunit constructs carrying the C351I mutation in (B, H) mGα_oA, (C, I) mGα_oB, (D, J) mGα_i1, (E, K) mGα_i2 or (F, L) mGα_i3. In the upper panels, cells were incubated with PTX (100 ng•mL⁻¹) for 16 h, and cAMP was production stimulated by forskolin (3 μM) followed by treatment with mINSL5 (100 nM). Results are normalized to the cAMP response stimulated by forskolin. In the lower panels, the ERK1/2 response was measured in cells incubated with PTX (100 ng•mL⁻¹) for 16 h followed by treatment with mINSL5 (100 nM). Results are expressed as percentage of response elicited by 10% FBS. Bars represent mean ± SEM of independent experiments [cAMP: n = 6 (mGα_oA, mGα_i3), n = 7 (pcDNA3, mGα_oB, mGα_i1, mGα_i2); ERK1/2: n = 5]. *P < 0.05; significantly different from the forskolin response (cAMP inhibition) or the baseline response (ERK1/2 activation) in PTX‐treated cells; repeated‐measures two‐way ANOVA followed by Dunnett's multiple comparisons test. ns, non‐significant.

Figure 4

Activation of RXFP4 receptors by hINSL5 or mINSL5 promotes increased cellular BrdU incorporation. Concentration–response relationships are shown for cellular BrdU incorporation stimulated by hINSL5 or mINSL5 (10⁻¹² to 10⁻⁸ M) in CHO‐RXFP4 cells. Results are expressed as percentage of response elicited by 10% FBS. Data points represent means ± SEM of 5 independent experiments. Veh, vehicle.

Figure 5

Real‐time kinetic BRET studies of interactions between RXFP4 receptors and (A) GRK2, (B) β‐arrestin 1, (C) β‐arrestin 2, (D) membrane‐bound KRas and (E) early endosome marker Rab5a induced by hINSL5 and mINSL5. CHO‐K1 cells were transiently co‐transfected with RXFP4‐Rluc8 and Venus‐tagged GRK2, β‐arrestin 1, β‐arrestin 2, Rab5a or KRas constructs. Following transfection, we stimulated the cells with hINSL5 or mINSL5 (200 nM). Ligand‐induced BRET ratios were calculated by subtracting the ratio for the vehicle‐treated sample from the BRET ratio for each ligand‐treated sample as described. Data points represent mean ± SEM of independent experiments (n = 5–8).

Figure 6

INSL5 inhibits glucose‐stimulated insulin secretion and calcium mobilisation in MIN6 murine insulinoma cells and cAMP accumulation in NCI‐H716 human enteroendocrine (NHI‐H716) cells. In (A), real‐time quantitative PCR (qPCR) assay to determine the expression levels of rxfp4 and glp1r that are expressed as ratios relative to actb, multiplied by 1000 (n = 6). In (B), MIN6 cells equilibrated for 2 h in KRB buffer, followed by stimulation in either 0 or 10 mM glucose by mINSL5 (0.1–200 nM) for 2 h. Supernatants were assayed for insulin (expressed in ng•mL⁻¹; n = 5). In (C), Ca^{2 +} mobilisation in MIN6 cells (n = 5) using Fluo‐4‐AM (1 μM) in response to 10 mM glucose alone or in the presence of mINSL5 (10⁻⁹ M) or GLP‐1 (10⁻⁸ M). Concentration–response relationships for inhibition of Ca^{2 +} responses (AUC) by INSL5 were bell shaped (D). In (E), NCI‐H716 human enteroendocrine cells stimulated with forskolin (3 μM) showed inhibition of cAMP accumulation with hINSL5 (10⁻⁷ M; n = 5). Data points represent mean ± SEM of n independent experiments. *P < 0.05; significantly different from control. RFU, Relative Fluorescence Unit ns, non‐significant.

Figure 7

RXFP4 receptor signal transduction pathways. Inhibitors used in studying RXFP4 receptor signalling mechanisms are boxed in red. Following INSL5 stimulation, RXFP4 receptors recruit multiple Gα_i/o subunits (predominantly Gα_oA, Gα_oB, Gα_i2) that activate a range of signalling pathways, including cAMP inhibition, ERK1/2 (Thr²⁰²/Tyr²⁰⁴), Akt (Thr³⁰⁸/Ser⁴⁷³) p38MAPK (Thr¹⁸⁰/Tyr¹⁸²) and S6RP (Ser²³⁵/²³⁶) via intermediary pathways such as mTORC1/C2, Src, PI3K and MEK1/2, leading to enhanced cell proliferation. Activated RXFP4 receptors interact with GRK2 and β‐arrestins (β‐Arr 1/2), which leads to movement of the receptor away from the plasma membrane into early endosome compartments. Note: speculative pathways are shown with dotted lines. mammalian target of rapamycin, mTOR; phosphoinositide‐dependent kinase, PDK.