Homologous and heterologous desensitization of guanylyl cyclase-B signaling in GH3 somatolactotropes

Abstract

The guanylyl cyclases, GC-A and GC-B, are selective receptors for atrial and C-type natriuretic peptides (ANP and CNP, respectively). In the anterior pituitary, CNP and GC-B are major regulators of cGMP production in gonadotropes and yet mouse models of disrupted CNP and GC-B indicate a potential role in growth hormone secretion. In the current study, we investigate the molecular and pharmacological properties of the CNP/GC-B system in somatotrope lineage cells. Primary rat pituitary and GH3 somatolactotropes expressed functional GC-A and GC-B receptors that had similar EC50 properties in terms of cGMP production. Interestingly, GC-B signaling underwent rapid homologous desensitization in a protein phosphatase 2A (PP2A)-dependent manner. Chronic exposure to either CNP or ANP caused a significant down-regulation of both GC-A- and GC-B-dependent cGMP accumulation in a ligand-specific manner. However, this down-regulation was not accompanied by alterations in the sub-cellular localization of these receptors. Heterologous desensitization of GC-B signaling occurred in GH3 cells following exposure to either sphingosine-1-phosphate or thyrotrophin-releasing hormone (TRH). This heterologous desensitization was protein kinase C (PKC)-dependent, as pre-treatment with GF109203X prevented the effect of TRH on CNP/GC-B signaling. Collectively, these data indicate common and distinct properties of particulate guanylyl cyclase receptors in somatotropes and reveal that independent mechanisms of homologous and heterologous desensitization occur involving either PP2A or PKC. Guanylyl cyclase receptors thus represent potential novel therapeutic targets for treating growth-hormone-associated disorders.

Fig. 1

Molecular and pharmacological characterization of natriuretic peptide receptors in GH3 somatolactotropes and primary rat pituitary tissue. a Total RNA was extracted from either GH3 cells or primary rat pituitaries, prior to cDNA synthesis and reverse transcription plus the polymerase chain reaction for the indicated gene targets. Data shown are representative of at least three independent experiments. b, c GH3 cells were stimulated with the indicated concentrations of (b) atrial natriuretic peptide (ANP) or (c) C-type natriuretic peptide (CNP) in physiological saline solution (PSS) containing 1 mM 3-isobutyl-1-methylxanthine (IBMX) for 15 min, before termination with 100 %(v/v) ice-cold ethanol. Following extraction under vacuum, total cGMP accumulation was measured by enzyme-immunoassay and concentration-response curves were constructed by using pre-existing equations in GraphPad Prism 5.0. The data shown are means ± SEM of three independent experiments, each performed in triplicate and are expressed as pmol/mg protein (n = 3)

Fig. 2

Rapid homologous desensitization of CNP-dependent but not ANP-dependent cGMP accumulation in GH3 cells. Cells were treated with PSS containing either (a) 100 nM CNP or (b) 100 nM ANP for 0–15 min in the presence of 1 mM IBMX. The data shown are means ± SEM of three independent experiments, each performed in triplicate and are expressed as pmol/ml per milligram protein (n = 3). The bar charts (right) indicate a comparison of the rate of cGMP accumulation between 0 and 5 min and between 5 and 15 min of stimulation (***P < 0.001, significantly different from the rate of cGMP accumulation between 0 and 5 min). c GH3 cells were treated as before with 100 nM CNP from 0 to 15 m in the presence or absence of either the protein phosphatase 2A (PP2A) inhibitor, okadaic acid (OA, 100 nM), the PP2B inhibitor, cypermethrin (1 nM), or the protein tyrosine phosphatase inhibitor, sodium orthovanadate (Na ₃ VO ₄, 1 mM) and assayed for total cGMP concentration. The data are means ± SEM of three independent experiments, each performed in triplicate and are expressed as the fold change in the rate of cGMP accumulation between 5 and 15 min (n = 3; **P < 0.01, cGMP accumulation maintained at a significant rate)

Fig. 3

Effect of chronic exposure to CNP and ANP on GC-B- and GC-A-dependent cGMP accumulation in GH3 cells. Cells were initially treated with (a) 100 nM CNP or (b) 100 nM ANP in PSS without IBMX for 0 to 6 h. The medium was removed and replaced with 100 nM CNP (a) or 100 nM ANP (b) for 15 min in the presence of 1 mM IBMX. The data shown are means ± SEM of three independent experiments, each performed in triplicate and are expressed as the percentage of the control (n = 3). ***P < 0.001, *P < 0.05, significantly different from the control. c, d Cells were initially treated with 0, 100 nM CNP, or 100 nM ANP for 6 h as before. All pre-treatments were removed and replaced either with (c) 0 or 100 nM CNP or with (d) 0 or 100 nM ANP, for 15 min in the presence of 1 mM IBMX. The data shown are means ± SEM of three independent experiments, each performed in triplicate and are expressed as the percentage of the control (n = 3). *P < 0.05, significantly different from the control

Fig. 4

CNP fails to alter expression of Npr2 mRNA or GC-B protein in GH3 somatolactotrope cells. a GH3 cells were stimulated with 0 or 100 nM CNP for 24 h prior to extraction of total RNA, cDNA synthesis, and quantitative PCR for endogenous Npr2 expression. The data shown are means ± SEM of three independent experiments (n = 3). b GH3 cells were stimulated with 100 nM CNP for the indicated time, prior to extraction of total protein and Western blot analysis for GC-B expression. Blots were stripped and re-probed for β-actin. Panels are representative of three independent experiments (n = 3)

Fig. 5

Lack of ligand-dependent internalization of GC-B or GC-A in GH3 somatolactotrope cells. GH3 cells were cultured on glass coverslips for 24 h prior to being treated with either 100 nM CNP for 0 (a), 1 h (b), or 4 h (c), or 100 nM ANP for 0 (d), 1 h (e), or 4 h (f). GC-A and GC-B immunoreactivity was detected by using Alexa488-conjugated secondary antibodies (green), whereas Alexa635-conjugated phalloidin was used to detect the actin cytoskeleton (red). Nuclear localization was detected by using 4,6-diamidino-2-phenylindole (DAPI) stain for orientation (blue). Panels are representative of three independent experiments (n = 3). Magnification ×40

Fig. 6

Evidence of sphingosine-mediated but not calcium-mediated, heterologous desensitization of GC-B signaling in GH3 somatolactotrope cells. GH3 cells were pre-treated with PSS containing either (a, b) 10 μM sphingosine-1-phosphate (S1P) or (c, d) 1 μM A23187 (calcium ionophore) for 30 min in the absence of IBMX, before a 15-min stimulation with either 100 nM CNP (a, c) or 100 nM ANP (b, d) in the presence of 1 mM IBMX. The data are means ± SEM of three independent experiments, each performed in triplicate and are expressed as pmol/mg protein (n = 3). **P < 0.01, significantly different from control response to CNP

Fig. 7

Protein kinase C (PKC)-dependent heterologous desensitization mediates the effects of thyrotrophin-releasing hormone (TRH) on CNP-stimulated GC-B signaling in GH3 somatolactotrope cells. a GH3 cells were pre-treated with PSS containing either 0 or 100 nM TRH, phorbol-12-myristate-13-acetate (PMA), or pituitary adenylate cyclase-activating polypeptide (PACAP) for 30 min in the absence of IBMX, before a 15-min stimulation with either 0 or 100 nM CNP in the presence of 1 mM IBMX. The data are means ± SEM of three independent experiments, each performed in triplicate and are expressed as % control reponse to CNP (n = 3). ***P < 0.001, significantly different from control response to CNP. b GH3 cells were pre-treated with PSS containing either 0 or 1 μM GF109203X for 30 min and a subsequent pre-treatment with 0 or 100 nM TRH in the continued absence or presence of GF109203X, before a 15-min stimulation with either 0 or 100 nM CNP in the presence of 1 mM IBMX. The data are means ± SEM of three independent experiments, each performed in triplicate and are expressed as % control reponse to CNP (n = 3). ***P < 0.001, significantly different from control response to CNP