Interaction of AIP with protein kinase A (cAMP-dependent protein kinase)

Interaction of AIP with protein kinase A (cAMP-dependent protein kinase) Abstract Germline mutations in the aryl hydrocarbon receptor-interacting protein (AIP) gene cause mostly somatotropinomas and/or prolactinomas in a subset of familial isolated pituitary adenomas (FIPA). AIP has been shown to interact with phosphodiesterases (PDEs) and G proteins, suggesting a link to the cyclic AMP (cAMP)-dependent protein kinase (PKA) pathway. Upregulation of PKA is seen in sporadic somatotropinomas that carry GNAS mutations, and those in Carney complex that are due to PRKAR1A mutations. To elucidate the mechanism of AIP-dependent pituitary tumorigenesis, we studied potential functional and physical interactions of AIP with PKA’s main subunits PRKAR1A (R1α) and PRKACA (Cα). We found that AIP physically interacts with both R1α and Cα; this interaction is enhanced when all three components are present, but maintained during Cα-R1α dissociation by PKA activation, indicating that AIP binds Cα/R1α both in complex and separately. The interaction between AIP and R1α/Cα is reduced when the frequent AIP pathogenic mutation p.R304* is present. AIP protein levels are regulated both by translation and the ubiquitin/proteasome pathway and Cα stabilizes both AIP and R1α protein levels. AIP reduction by siRNA leads to an increase of PKA activity, which is disproportionately enhanced during PDE4-inhibition. We show that AIP interacts with the PKA pathway on multiple levels, including a physical interaction with both the main regulatory (R1α) and catalytic (Cα) PKA subunits and a functional interaction with PDE4-dependent PKA activation. These findings provide novel insights on the mechanisms of AIP-dependent pituitary tumorigenesis. Introduction Germline mutations in the aryl hydrocarbon receptor interacting protein (AIP, OMIM*605555) have recently been identified as the genetic cause of ∼20% of familial isolated pituitary adenomas (FIPA, OMIM#102200) (1–3). AIP-associated pituitary adenomas are mostly early-onset, aggressive somatotropinomas, while patients demonstrate no other recurrent clinical manifestations (2–4). The molecular mechanism by which these inactivating AIP mutations lead to pituitary tumor formation is currently unclear. AIP interacts with components of the cyclic adenosine monophosphate (cAMP)/protein kinase A (PKA) pathway, including the phosphodiesterases (PDEs) 4A5 and 2A3, as well as the G proteins Gα13 and Gαq (5–9). Functionally, a reduction of Aip levels in GH3 rat pituitary somatotroph cells led to an increase in PKA signaling on different levels (10); however, this effect was not completely abolished by PDE inhibition (10), suggesting that other components of the PKA pathway mediate AIP-dependent effects on pituitary tumor development. The cAMP/PKA pathway has been shown to be involved in the regulation of pituitary cell proliferation and pituitary hormone secretion and it is frequently dysregulated in pituitary tumors (11,12). Approximately 40% of sporadic somatotropinomas were shown to carry somatic mutations of the α-subunit of the Gs protein (GNAS), leading to adenylyl cyclase and hence PKA pathway activation (13–16). McCune-Albright syndrome (OMIM#174800) is caused by a postzygotic mutation of GNAS and ∼20% of patients present with somatotroph or somato-lactotroph hyperplasia and growth hormone (GH) hypersecretion (17,18). Carney complex (CNC, OMIM#160980)-associated somatotropinomas are caused by inactivating mutations of the PKA regulatory subunit 1α (PRKAR1A), leading to a lack of inhibition of the PKA catalytic subunits, mainly Cα (19,20). Even in the absence of PRKAR1A mutations or loss of heterozygosity, R1α protein expression is markedly reduced in sporadic pituitary tumors including somatotropinomas, which contributes to somatotroph proliferation (11). In the present investigation, we examined the hypothesis that AIP may functionally and physically interact with components of the PKA pathway, including the main PKA regulatory (R1α) and catalytic (Cα) subunit, encoded, respectively, by the PRKAR1A and PRKACA genes. The data uncover new AIP interactions that point to cAMP/PKA’s involvement in the former’s tumor suppressor role. Results Intracellular localization and co-localization of AIP, R1α and Cα Confocal immunofluorescence shows that endogenous AIP colocalizes with R1α and Cα in the rat mammosomatotropinoma cell line GH3 (Fig. 1A and B). Colocalization was observed mainly in cytoplasmic foci, potentially relating to the cytoplasmic anchoring of R1α to A-kinase anchoring proteins (AKAPs) (21). The colocalization between AIP and Cα significantly increases during PKA activation by forskolin (fsk) compared to vehicle (Pearson’s R above threshold 0.765 versus 0.721, P = 0.012). Fsk, which is widely used to stimulate pituitary hormone secretion in vitro, activates adenylyl cyclase, raises intracellular cAMP levels and thereby activates PKA (10,22). Colocalization decreases in the presence of the PKA inhibitor H-89 (Pearson’s R above threshold 0.693, P = 0.006, Supplementary Material, Fig. S1). AIP colocalization with R1α is not significantly affected by fsk or H-89 (Supplementary Material, Fig. S1). Figure 1. View largeDownload slide Intracellular localization and colocalization of AIP, Cα and R1α. (A and B) Confocal immunofluorescence of AIP (green channel) with Cα (red channel) (A) or with R1α (B) in GH3 cells at 63× magnification. Nuclei were visualized with DAPI (blue channel). Cells were treated with 10 µM forskolin (fsk) or DMSO (vehicle) for 30 min. (C) GH3 cells overexpressing AIP (AIP +) or empty vector (AIP−) were subjected to cell fractionation to separate cytosolic (Cyt), membranous (mem), nuclear and nuclear bound (N. b.) compartments. Fractions were subjected to Western blotting and R1ɑ, Cα and AIP as well as GAPDH, GR, e-cadherin, histone H3, HPRT and TORC were detected. Figure 1. View largeDownload slide Intracellular localization and colocalization of AIP, Cα and R1α. (A and B) Confocal immunofluorescence of AIP (green channel) with Cα (red channel) (A) or with R1α (B) in GH3 cells at 63× magnification. Nuclei were visualized with DAPI (blue channel). Cells were treated with 10 µM forskolin (fsk) or DMSO (vehicle) for 30 min. (C) GH3 cells overexpressing AIP (AIP +) or empty vector (AIP−) were subjected to cell fractionation to separate cytosolic (Cyt), membranous (mem), nuclear and nuclear bound (N. b.) compartments. Fractions were subjected to Western blotting and R1ɑ, Cα and AIP as well as GAPDH, GR, e-cadherin, histone H3, HPRT and TORC were detected. In cell fractionation experiments, AIP is primarily localized within the cytoplasm and in the membranous fraction (Fig. 1C). Similarly, R1α and Cα mainly localize to the cytoplasm and membranes; a small amount of Cα is detectable in the unbound nuclear fraction. The purity of fractions and equal loading were verified by visualizing the expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH, expressed in the cytosol and perimembranous), the glucocorticoid receptor (GR, expressed in the cell membrane and in the nucleus), e-cadherin (cell membrane expression), histone H3 (chromatin-bound in the nuclear bound fraction), hypoxanthine-guanine phosphoribosyltransferase (HPRT, cytosolic expression) and the transcriptional co-activator transducer of regulated CREB-binding proteins (TORC, chromatin-bound). AIP overexpression leads to a decrease in membranous R1α levels as well as a decrease in nuclear Cα expression, compared to control (empty vector). The specificity of the AIP antibody was verified during AIP overexpression in a separate experiment (Supplementary Material, Fig. S2). Physical interaction between AIP, R1α and Cα The influence of AIP on the interaction between R1α and Cα was tested in fluorescence resonance energy transfer (FRET) by acceptor photobleaching. R1α-Venus and Cα-Cerulean were co-overexpressed with or without AIP, and their interaction was measured by the increase of Cerulean fluorescence intensity after photobleaching Venus. Forskolin treatment caused a decrease in donor intensity to 42% (P < 0.001) relative to control, consistent with dissociation and decreased interaction between R1α and Cα after PKA pathway activation (Fig. 2A). AIP overexpression showed a tendency towards a decrease in R1α-Cα interaction in basal conditions, although this did not reach statistical significance. In a two-way ANOVA an interaction effect was detected, suggesting a synergistic effect of Fsk and AIP overexpression in the dissociation of the PKA subunits (P = 0.002 for interaction effects between Fsk*AIP overexpression in two-way ANOVA). Figure 2. View largeDownload slide Physical interaction between AIP and the PKA. (A) GH3 cells were co-transfected with R1α-Venus and Cα-Cerulean as well as either empty vector (EV) or AIP. After 24 h, cells were stimulated with 10 µM forskolin (Fsk) or vehicle for 1 h before imaging. The interaction between R1α and Cα was then quantified via fluorescence resonance energy transfer (FRET) by acceptor (R1α-Venus) photobleaching. Differences between means were tested by two-way ANOVA followed by post hoc testing. ***P < 0.001, n.s. not significant. Data are expressed relative to EV/vehicle control and represent the mean of three separate experiments. (B) Co-IP of AIP-Myc on Myc-coupled resin with R1α and HSC70. AIP-Myc, HSC70-HA and R1α-HA were overexpressed; IPs were performed with or without crosslinking with paraformaldehyde (PFA). (C) Co-IP of AIP-Myc on Myc-coupled resin with R1α and HSP90. AIP-Myc, HSP90-HA and R1α-HA were overexpressed; IPs were performed with or without crosslinking with paraformaldehyde (PFA). (D) Co-IP of R1α-HA on HA-coupled resin with AIP-Myc. R1α-HA and AIP-Myc were overexpressed alone or in combination (top panel). IgG alone was run in the left lane of each Western blot, while the right two lanes show the respective controls with IgG-coupled resin. (E) Co-IP of AIP-Myc on Myc-coupled resin with R1α-HA and Cα-HA. Different combinations of R1α-HA, Cα-HA and AIP-Myc were overexpressed (top panel). The left two lanes were additionally stimulated with 10 µM forskolin (Fsk). (F) Co-IP of Cα-HA on HA-coupled resin with AIP-Myc and R1α. Cα-HA, R1α-pcDNA and either AIP-Myc wild-type (AIP wt) or AIP-Myc R304* (AIP R304*) were overexpressed (top panel). (G) Co-IP of R1α-HA on HA-coupled resin with AIP-Myc and Cα. R1α-HA, Cα-pCMV and either AIP-Myc wild-type (AIP wt) or AIP-Myc R30`4* (AIP R304*) were overexpressed (top panel). Figure 2. View largeDownload slide Physical interaction between AIP and the PKA. (A) GH3 cells were co-transfected with R1α-Venus and Cα-Cerulean as well as either empty vector (EV) or AIP. After 24 h, cells were stimulated with 10 µM forskolin (Fsk) or vehicle for 1 h before imaging. The interaction between R1α and Cα was then quantified via fluorescence resonance energy transfer (FRET) by acceptor (R1α-Venus) photobleaching. Differences between means were tested by two-way ANOVA followed by post hoc testing. ***P < 0.001, n.s. not significant. Data are expressed relative to EV/vehicle control and represent the mean of three separate experiments. (B) Co-IP of AIP-Myc on Myc-coupled resin with R1α and HSC70. AIP-Myc, HSC70-HA and R1α-HA were overexpressed; IPs were performed with or without crosslinking with paraformaldehyde (PFA). (C) Co-IP of AIP-Myc on Myc-coupled resin with R1α and HSP90. AIP-Myc, HSP90-HA and R1α-HA were overexpressed; IPs were performed with or without crosslinking with paraformaldehyde (PFA). (D) Co-IP of R1α-HA on HA-coupled resin with AIP-Myc. R1α-HA and AIP-Myc were overexpressed alone or in combination (top panel). IgG alone was run in the left lane of each Western blot, while the right two lanes show the respective controls with IgG-coupled resin. (E) Co-IP of AIP-Myc on Myc-coupled resin with R1α-HA and Cα-HA. Different combinations of R1α-HA, Cα-HA and AIP-Myc were overexpressed (top panel). The left two lanes were additionally stimulated with 10 µM forskolin (Fsk). (F) Co-IP of Cα-HA on HA-coupled resin with AIP-Myc and R1α. Cα-HA, R1α-pcDNA and either AIP-Myc wild-type (AIP wt) or AIP-Myc R304* (AIP R304*) were overexpressed (top panel). (G) Co-IP of R1α-HA on HA-coupled resin with AIP-Myc and Cα. R1α-HA, Cα-pCMV and either AIP-Myc wild-type (AIP wt) or AIP-Myc R30`4* (AIP R304*) were overexpressed (top panel). Co-immunoprecipitation (co-IP) experiments were conducted to further clarify a potential interaction of AIP with R1α or Cα. Since AIP is bound and stabilized by the two chaperones HSP90 (heat shock protein 90) and HSC70 (heat shock cognate 70), which assist in its folding and may facilitate protein–protein interactions (5), we suspected that any AIP-PKA interaction may also be stabilized by the presence of HSP90 or HSC70. We found that AIP interacts with R1α both in the presence (Fig. 2B and C) and absence (Fig. 2D) of HSC70 and HSP90, thus indicating that the AIP-R1α interaction is independent of the presence of these two chaperone proteins. In fact, the presence of HSP90 does not appear to increase the amount of R1α bound to AIP (data not shown). AIP also interacts with Cα: this interaction is increased in the presence of R1α, suggesting that an AIP-R1α-Cα complex may be a more stable configuration. Furthermore, this interaction is maintained in the presence of the PKA activator forskolin (Fig. 2E), which leads to the dissociation of the subunits, thereby suggesting that AIP can bind R1α and Cα in their inactive complex as well as the active separate subunits. Conversely, the influence of Cα on the AIP-R1α association cannot be determined since AIP and R1α protein expression levels are also highly dependent on the presence of Cα (Fig. 4). The interaction between AIP and Cα (Fig. 2F) as well as that of AIP and R1α (Fig. 2G) is reduced in the truncated AIP R304* mutant that lacks most of the C-terminal α-7 helix that is crucial for protein–protein interactions. Levels of AIP co-IP remain reduced even when correcting for the lower expression of AIP R304* compared to wild-type (R1α-AIP R304* 0.5-fold, Cα-AIP R304* 0.33-fold compared to wild-type), suggesting that the α-7 helix of AIP is involved in the interaction between AIP and R1α/Cα, but other parts of AIP also contribute to this interaction. AIP reduces PKA activity Basal and total PKA activity (in the absence or presence of cAMP, respectively) were determined in GH3 cells during AIP overexpression. While no significant difference was detected in basal PKA activity between AIP and control (Fig. 3A), total PKA activity was significantly lower during AIP overexpression (Fig. 3B). We did not observe a significant effect of AIP silencing by siRNA on basal or total PKA activity (data not shown). Figure 3. View largeDownload slide PKA activity in response to AIP overexpression. AIP or empty vector (EV) were overexpressed in GH3 cells for 48 h; cells were lysed and PKA activity was determined in the absence [basal PKA activity (A)] and in the presence [total PKA activity (B)] of cAMP. Differences between means were tested by independent samples Student’s t-tests. ***P < 0.001, n.s. not significant. Data shown are expressed relative to EV and represent the mean from two separate experiments. Figure 3. View largeDownload slide PKA activity in response to AIP overexpression. AIP or empty vector (EV) were overexpressed in GH3 cells for 48 h; cells were lysed and PKA activity was determined in the absence [basal PKA activity (A)] and in the presence [total PKA activity (B)] of cAMP. Differences between means were tested by independent samples Student’s t-tests. ***P < 0.001, n.s. not significant. Data shown are expressed relative to EV and represent the mean from two separate experiments. Cα stabilizes AIP and R1α protein levels AIP protein levels are increased in the presence of Cα (Fig. 4A and D). During overexpression, AIP protein levels are ∼2-fold increased when Cα is co-overexpressed. The presence of Cα also increases R1α levels by >10-fold (Fig. 4A and D). The stabilizing effect of Cα on AIP protein levels is lightly dampened by the additional presence of R1α, which may limit Cα binding capacities to AIP. Aip mRNA levels do not significantly differ in the presence of Cα (Fig. 4B), indicating that Cα stabilizes AIP and R1α only at the protein level. Aip mRNA levels are similarly not affected during activation of the PKA pathway with fsk and consequent dissociation of the PKA subunits or during PKA pathway inhibition (when PKA subunits are in complex) with the PKA inhibitor peptide PKI (Fig. 4C). These results suggest that the stabilizing effect of Cα on AIP is independent of PKA pathway activity. Figure 4. View largeDownload slide Regulation of AIP mRNA and protein levels by Cα and R1α. (A) GH3 cells were co-transfected with different combinations of AIP, Cα and R1α (as indicated, top panel). Total cell lysate was subjected to Western blotting. Expression levels were compared to tubulin as a loading control. (B) GH3 cells were co-transfected with Cα and/or R1α, as indicated. Expression levels of Aip mRNA were determined by RT-qPCR in relation to β-Actin levels. (C) GH3 cells were treated with 10 µM forskolin (Fsk) or 1 µM PKA inhibitor (PKI) for 8 h. Aip expression levels were determined by RT-qPCR in relation to β-Actin levels. (D) GH3 cells were co-transfected with different combinations of AIP, Cα and R1α (as indicated, top panel), followed by treatment with 5 µg/ml cycloheximide (CHX), 1 µM MG132 or vehicle. Total cell lysate was subjected to Western blotting. Expression levels were compared to tubulin as a loading control. Differences between means were tested by one-way ANOVA followed by post hoc testing (B and C). n.s. not significant. Data are expressed relative to EV (B) or vehicle (C) controls and represent the mean from one representative experiment performed in triplicate wells. Figure 4. View largeDownload slide Regulation of AIP mRNA and protein levels by Cα and R1α. (A) GH3 cells were co-transfected with different combinations of AIP, Cα and R1α (as indicated, top panel). Total cell lysate was subjected to Western blotting. Expression levels were compared to tubulin as a loading control. (B) GH3 cells were co-transfected with Cα and/or R1α, as indicated. Expression levels of Aip mRNA were determined by RT-qPCR in relation to β-Actin levels. (C) GH3 cells were treated with 10 µM forskolin (Fsk) or 1 µM PKA inhibitor (PKI) for 8 h. Aip expression levels were determined by RT-qPCR in relation to β-Actin levels. (D) GH3 cells were co-transfected with different combinations of AIP, Cα and R1α (as indicated, top panel), followed by treatment with 5 µg/ml cycloheximide (CHX), 1 µM MG132 or vehicle. Total cell lysate was subjected to Western blotting. Expression levels were compared to tubulin as a loading control. Differences between means were tested by one-way ANOVA followed by post hoc testing (B and C). n.s. not significant. Data are expressed relative to EV (B) or vehicle (C) controls and represent the mean from one representative experiment performed in triplicate wells. To investigate the cellular mechanisms governing AIP, R1α and Cα protein levels, experiments were conducted using the inhibitor of protein synthesis cycloheximide (CHX) and the proteasome inhibitor MG132. In the absence of R1α and Cα overexpression, CHX leads to a reduction of overexpressed AIP levels within 14 h, while the addition of MG132 rescues this effect, indicating that AIP protein levels are regulated by translation and degradation via the ubiquitin/proteasome pathway (Fig. 4D). R1α levels are increased after proteasome inhibition, whereas Cα levels are more strongly influenced by translational inhibition. When all three AIP, R1α and Cα are co-overexpressed there appears to be a stronger overall effect of CHX than MG132, indicating that the three proteins together are strongly regulated by translation (Fig. 4D). AIP functionally interacts with PDE-dependent PKA pathway activity via PDE4 AIP is known to directly interact with PDE2 and PDE4 (5), while genetic variants of PDE11A were recently investigated in the context of acromegaly (23). We first tested the effect of different subtype-specific PDE inhibitors on cAMP-responsive element (CRE)-activation in a double luciferase assay in GH3 cells. The PDE2 inhibitors EHNA and BAY 60–7550, and the PDE11A inhibitor BC11–38 did not have an effect on CRE-activity, while the pan-PDE inhibitor IBMX and the PDE4 inhibitor Rolipram increased basal CRE-activity (Fig. 5A), suggesting a physiologically important role of PDE4 in PKA signaling in pituitary somatotrophs (24,25). Figure 5. View largeDownload slide Functional interaction between AIP and PDE-dependent PKA pathway activity. (A) GH3 cells were co-transfected with a cAMP responsive element luciferase reporter vector (CRE-Luc) and a control Renilla luciferase vector (R-Luc). The following day, cells were stimulated with different PDE inhibitors or vehicle, as stated, for 8 h, after which CRE-Luc/R-Luc were assessed in a double luciferase assay. (B) GH3 cells were co-transfected with CRE-Luc, R-Luc and Aip siRNA or control siRNA. The following day, cells were stimulated with either IBMX, Rolipram or vehicle for 8 h, followed by a double luciferase assay. Differences between means were tested by one-way ANOVA (A) or two-way ANOVA (B) followed by post hoc testing. *P < 0.05, ***P < 0.001, n.s. not significant. Data shown are expressed relative to vehicle (A) or vehicle/EV (B) controls and represent the mean from one representative experiment performed in triplicate wells. Figure 5. View largeDownload slide Functional interaction between AIP and PDE-dependent PKA pathway activity. (A) GH3 cells were co-transfected with a cAMP responsive element luciferase reporter vector (CRE-Luc) and a control Renilla luciferase vector (R-Luc). The following day, cells were stimulated with different PDE inhibitors or vehicle, as stated, for 8 h, after which CRE-Luc/R-Luc were assessed in a double luciferase assay. (B) GH3 cells were co-transfected with CRE-Luc, R-Luc and Aip siRNA or control siRNA. The following day, cells were stimulated with either IBMX, Rolipram or vehicle for 8 h, followed by a double luciferase assay. Differences between means were tested by one-way ANOVA (A) or two-way ANOVA (B) followed by post hoc testing. *P < 0.05, ***P < 0.001, n.s. not significant. Data shown are expressed relative to vehicle (A) or vehicle/EV (B) controls and represent the mean from one representative experiment performed in triplicate wells. To test a possible functional interaction between AIP and PDEs in GH3 cells, Aip reduction by siRNA-mediated knockdown and PDE inhibition were performed concurrently. Aip knockdown led to a modest increase of PKA activity as measured by CRE-activity (Fig. 5B). IBMX and Rolipram also lead to increased CRE-activation, which is even further increased in the absence of Aip. Levels of CRE-activation were compared during Aip siRNA, PDE inhibition by IBMX and Rolipram, and both simultaneous treatments in a two-way ANOVA. We found a significant interaction effect between Aip knockdown and PDE inhibition (P < 0.001 for interaction effects between Aip siRNA*IBMX and P < 0.001 for interaction effects between Aip siRNA*Rolipram); the effect of Aip siRNA on CRE-activation is proportionately higher during PDE-inhibition by IBMX and Rolipram compared to control [1.45-fold (control) versus 2.37-fold increase for IBMX and 1.75-fold increase for Rolipram]. This suggests that AIP functionally interacts with PDE activity in pituitary somatotrophs, and this interaction is at least partially mediated via PDE4. Discussion We show here that there is a cytoplasmic and perimembranous colocalization of AIP with the regulatory and catalytic subunits of the PKA, whereby AIP-Cα colocalization increases during PKA pathway activation. AIP physically interacts with both R1α and Cα. This interaction is enhanced when all three components are present, but maintained during Cα-R1α dissociation by PKA activation, indicating that AIP binds Cα and R1α both in complex and separately. AIP overexpression leads to a decrease of PKA activity. The interaction between AIP and R1α/Cα is reduced in the frequent pathogenic mutant AIP p.R304*, which lacks its C-terminal α-helix. AIP protein levels are regulated by translation and the ubiquitin/proteasome pathway and Cα stabilizes both AIP and R1α protein levels. Aip reduction by siRNA leads to an increase of PKA pathway activity, which is disproportionately enhanced during PDE4-inhibition, indicating that Aip functionally interacts with PDE4-dependent PKA pathway activity. While it is known that AIP deficiency leads to altered PKA signaling in pituitary somatotrophs (10), a physical interaction between AIP and the PKA was not previously demonstrated. We show here that AIP interacts with R1α as well as Cα. Conversely, the catalytic PKA subunit stabilizes AIP protein levels. While we cannot exclude that these interactions are mediated via yet another mutual interaction partner (26), we provide evidence that both these interactions are maintained during PKA activation when both PKA subunits are separate. AIP-Cα binding may influence Cα function when AIP is bound to Cα. We show that Aip-Cα colocalization is increased during PKA activation and the dissociation of regulatory and catalytic PKA subunits. The observed cytoplasmic colocalization between AIP and R1α as well as Cα was not uniform but confined to some cytoplasmic areas. The intracellular localization of the PKA is known to be regulated via the AKAPs which act as a scaffold for the PKA regulatory and catalytic subunits as well as PKA substrates and other components of the PKA pathway, thereby regulating PKA pathway activity and specificity (21). AIP was also previously shown to localize to secretory granules in somatotrophs (27). Furthermore, we show that AIP overexpression altered nuclear Cα levels, suggesting that AIP-Cα binding may modify Cα localization and hence its activity; this could be another mechanism by which AIP deficiency may contribute to pituitary tumor formation. Functionally, AIP overexpression led to a decrease in total PKA activity, and additionally showed a trend towards lowering basal PKA activity. These findings are in agreement with the observation that AIP binds to the PKA subunits in the inactive as well as in the active configuration. Mechanistically, AIP may bind Cα and thereby inhibit its catalytic activity, or it may bind the inactive Cα-R1α complex and prevent subunit dissociation, thereby decreasing levels of active Cα. We and others (10) found that Aip silencing leads to an increase in PKA pathway activation as measured by CRE-activity (Fig. 5B), while basal and total PKA activity were not significantly changed, overall suggesting that Aip silencing causes a modest increase in PKA pathway activity that is not due to direct activation of the PKA. The truncating AIP p.R304* mutant, one of the most common mutations detected in human AIP-dependent FIPA patients, interacts with the PKA subunits to a lesser degree than the wild-type, suggesting that the binding between the PKA subunits and AIP may contribute to the pathogenic mechanism of human AIP-dependent pituitary tumorigenesis. Furthermore, these results demonstrate that the C-terminal α-7 helix of AIP, which is truncated in AIP p.R304*, is involved in but not solely responsible for AIP-R1α and AIP-Cα binding. To further elucidate the importance of different regions of AIP in the interaction with the PKA, and to infer the contribution of this interaction to pituitary tumorigenesis, other pathogenic mutants should be investigated in the future. AIP contains three tetratricopeptide repeat (TPR) domains, which were shown to contribute to AIP homodimerization as well as to AIP binding to a plethora of interacting partners including PDE4A5 and HSP90 (5,28). In addition, many pathogenic AIP missense mutations are localized within the TPR domains, thereby highlighting their importance in AIP-dependent pituitary tumorigenesis (5). These TPR domains could equally contribute to the binding of AIP to R1α/Cα. While AIP binds both R1α and Cα, the binding affinities to each PKA subunit is likely to be different and may contribute to potential different functional effects of the AIP-R1α and the AIP-Cα interaction. Considering this, intracellular AIP levels and hence the proportion of AIP bound to Cα or R1α or the Cα-R1α complex may be crucial in the contribution of AIP to PKA pathway activity. The reduction of half-life of AIP mutants was recently shown to be a major contributor to the pathogenic mechanism in pituitary tumorigenesis (29). We confirm and extend these findings by demonstrating that 1. AIP protein levels but not mRNA levels are strongly influenced by Cα, and to a lesser degree, R1α; 2. both Cα and R1α physically bind to AIP; 3. this binding is reduced in the AIP p.R304* mutant, which was also shown to have a severely reduced half-life (29). We propose that the alterations in AIP half-life may at least partially be mediated by impaired AIP-Cα/R1α binding, and that a reduction of this binding contributes to pituitary tumorigenesis by modifying PKA pathway activity. Our data also confirm a role of the ubiquitin proteasome pathway in the regulation of intracellular AIP levels, in addition to translational control. Either or both of those mechanisms may be responsible for the stabilizing effect of Cα (and R1α to a smaller extent) on cellular AIP levels. The intracellular action of AIP in the context of pituitary tumor formation is complex and other mechanisms are also likely to contribute to AIP-dependent pituitary tumorigenesis. Others have shown that AIP interacts with the inhibitory pathway involving Gαi (30). AIP is known to bind to PDEs (5); AIP binds and inhibits PDE4A5 catalytic activity via its TPR domain (6,9). Our data demonstrate that reduced AIP levels disproportionately enhance PKA pathway activity under PDE4-specific inhibition in pituitary somatotrophs; hence disturbance of the AIP-PDE4 interaction may contribute to AIP-dependent pituitary tumorigenesis. AIP is a member of the aryl hydrocarbon receptor (AHR) signaling pathway. The AHR binds environmental toxins and mediates the cellular transcriptional response to those toxins; however the AHR pathway is also involved in physiologically essential ligand-independent functions, including the regulation of cell growth and differentiation (31). Nuclear translocation of AHR is stimulated by cAMP (32), demonstrating an overlap between the AHR and the PKA pathways, which may form the basis of AHR’s ligand-independent functions. Since AIP also acts on cAMP levels (10), AIP could potentially be an upstream regulator of ligand-independent AHR effects in addition to PKA pathway activity; further studies are necessary to determine if AIP can indirectly affect AHR translocation via altered cAMP levels. The interaction between AIP and the PKA we show here present an additional point of overlap between the AHR and PKA pathways. In a pituitary-specific context, decreased expression of AHR and AHR nuclear translocator (ARNT) were described in AIP-dependent pituitary adenomas (33–35). Furthermore, recent data also demonstrated that transcriptional targets of the AHR pathway were altered in fibroblasts of patients carrying inactivating AIP mutations (36). R1α and Cα are regarded as the main PKA subunits; they are the most highly expressed and best studied regulatory and catalytic PKA subunits (37). A low ratio between R1α and the R2 subunits (R2α and R2β) by regulation of R1α levels in human somatotropinomas was suggested as a potential mechanism contributing to pituitary tumorigenesis (11). The role of different PKA catalytic subunits in the pituitary has not been investigated in detail; in addition to the interaction between R1α and Cα with AIP shown here, other PKA subunits may also play a role in AIP-dependent PKA pathway modulation and this should be investigated in the future. We propose that the mechanisms of AIP-dependent pituitary tumor formation are complex and involve an impact on PKA pathway activation upstream of the PKA, leading to quantitative changes of PKA activity, as well as a physical interaction with PKA subunits governing downstream effects. While this study focused primarily on the mechanisms of AIP towards pituitary tumorigenesis, AIP is an ubiquitously expressed protein whose function(s) in other tissues is still not well defined. Since both the PKA pathway and the AHR pathway are ubiquitous regulators of metabolism and proliferation, an interaction between AIP and the PKA pathway is also highly relevant for other areas of research and tumor entities. We identified here a physical interaction between AIP and the PKA pathway and its subunits, which may play an important role for pituitary tumorigenesis. A clear understanding of the molecular mechanisms of tumorigenesis is crucial for the development of novel rational therapeutic strategies. Materials and Methods Cell culture, plasmids and transfection The rat mammosomatotropinoma cell line GH3 and the human embryonic kidney cell line HEK293 were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and cells were maintained under standard cell culture conditions in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/ml penicillin, 100 μg/ml streptomycin and 0.25 μg/ml amphotericin B (all from Thermo Fisher Scientific, Waltham, MA). For transfections, cells were seeded into 12-well plates [real-time quantitative PCR (RT-qPCR) and Western blotting] or 10 cm dishes (cell fractionation) and left overnight to recover. The following day, 800 ng (for 12-well plates) of plasmid or empty vector (EV) were transfected into cells using Lipofectamine 2000 (Thermo Fisher Scientific) according to manufacturer’s recommendations. Effects on transcription and translation were assessed after 24 h unless stated otherwise. The plasmids used were: AIP-Myc-pcDNA, AIP_R304*-Myc-pcDNA (27), pcDNA-PRKAR1A (38), pCMV-PRKACA (OriGene, Rockville, MD), PRKAR1A-HA, PRKACA-HA, HSC70-HA and HSP90-HA [all cloned into the pSF-CMV-NH2-HA-EKT-NcoI vector (Oxford Genetics OG93, Oxford, UK)], pVenus-PRKAR1A, pCerulean-PRKACA (38), a firefly luciferase reporter vector driven by a cAMP responsive element (CRE, pGL4.29[luc2P/CRE/Hygro], Promega, Madison, WI) and a Renilla luciferase reporter control vector (pRL-SV40, Promega). For siRNA experiments, 2 μl of 20 μM Aip siRNA or non-targeting control siRNA (Thermo Fisher Scientific) were transfected analogous to overexpression experiments (above, same paragraph). For stimulations, cells were seeded into 12-well plates and left overnight to recover. Cells were then stimulated with PKI (PKA inhibitor 6–22 amide, Merck Millipore, Darmstadt, Germany) or forskolin (fsk, Sigma-Aldrich, St. Louis, MO) for 8 h. Cells were then lysed and subjected to RNA isolation. For protein stability experiments, cells were seeded into 12-well plates and left overnight to recover. Cells were then transfected and left to recover for 6 h, before placing them in serum-free medium. After 4 h, the translation inhibitor cycloheximide (CHX, 5 μg/ml, Sigma-Aldrich), and the proteasome inhibitor MG132 (1 μM, Sigma-Aldrich) or vehicle (dimethyl sulfoxide, DMSO, Sigma-Aldrich) were added. After an incubation period of 14 h, cells were lysed and subjected to Western blotting. RNA isolation, reverse transcription and RT-qPCR RNA was isolated using the RNeasy kit (Qiagen) including on-column DNase I treatment. About 500 ng of RNA were reverse transcribed using Superscript III (Life Technologies) and cDNA levels were quantified by multiplex RT-qPCR using TaqMan gene expression assays for rat Aip (Rn00597273_m1) along with beta-Actin (Actb, Rn00667869_m1) on a ViiA Real-Time PCR system (all from Life Technologies) according to manufacturer’s instructions. Relative gene expression was calculated using the ΔΔCt method. Confocal immunofluorescence GH3 cells were seeded onto poly-l-lysine (Sigma-Aldrich) coated glass chamber slides (Thermo Fisher Scientific) and left overnight to recover. They were then placed in low-serum medium (0.5% FBS) for 24 h before stimulations. Cells were pre-incubated with the competitive PKA inhibitor H-89 (Sigma-Aldrich) at 10 μM or vehicle for 1 h before stimulation with the adenylyl cyclase activator forskolin (fsk, Sigma-Aldrich) at 10 μM or vehicle for 30 min. Cells were then washed twice with phosphate-buffered saline (PBS, Thermo Fisher Scientific) and fixed in 4% paraformaldehyde (Sigma-Aldrich) for 15 min. They were blocked in 10% horse serum (Jackson ImmunoResearch Labs, West Grove, PA) in phosphate-buffered saline (PBS) containing 0.1% Triton X-100 (Sigma-Aldrich) for 1 h and then incubated with primary antibodies anti-AIP (1:200, Novus Biologicals, Littleton, CO) and anti-PRKAR1A (1:100, Abcam, Cambridge, UK) or anti-PRKACA (1:50, Santa Cruz Biotechnology, Dallas, TX) in 10% horse serum for 2 h at room temperature. Slides were washed and incubated with secondary antibodies Alexa Fluor 488 donkey anti-mouse and Alexa Fluor 555 donkey anti-rabbit (1:2000, Thermo Fisher Scientific) in PBS for 1 h at room temperature. Slides were washed and mounted using Prolong Gold mount with DAPI (4′, 6-diamidino-2-phenylindole, Thermo Fisher Scientific). Negative controls without primary antibody were included in every experiment. Confocal immunofluorescence experiments were performed as described previously (38). Briefly, images were captured using a Zeiss LSM 510 inverted confocal microscope with a 63× Zeiss plan-apochromat oil immersion objective (NA 1.4) or a 40× Zeiss plan-neofluar oil immersion objective (NA 1.3), as stated, and a 488-nm Argon ion laser with a 505–530 nm band pass emission filter, a 543-nm helium-neon laser and a 560–615 emission filter, and a 405-nm solid-state laser with a 420–470 nm emission filter, with the LSM 510 software. Colocalization was quantified in images taken from 10 separate visual fields using the Coloc2 plugin in ImageJ (National Institutes of Health, Bethesda, MD) to calculate Pearson’s R above threshold. Differences between means were assessed by independent samples Student’s t-tests. Fluorescence resonance energy transfer by acceptor photobleaching GH3 cells were seeded onto poly-l-lysine (Sigma-Aldrich) coated glass chamber slides (Thermo Fisher Scientific) and left overnight to recover. pVenus-R1α (acceptor), pCerulean-Cα (donor) and either AIP or EV were transfected and left for 24 h. About 1 h prior to FRET, cells were stimulated with either fsk or vehicle. Experiments were performed on the same microscope and with the same software used for confocal immunofluorescence (see above). Cα-Cerulean was imaged with the 405-nm laser, R1α-Venus with the 514-nm laser. Using a custom region of interest, R1a-Venus in one cell was bleached with the 514-nm laser at 100% transmission until the overall intensity dropped to between 80% and 50% of pre-bleach values. The intensity of Cα-Cerulean and R1α-Venus were concurrently measured in a neighboring cell not exposed to the bleaching laser to ensure that no significant bleaching of either proteins took place during imaging. The R1α-Cα interaction was calculated by the difference in Cerulean intensity pre- versus post-bleaching. In each experiment, 15–20 bleaches on different cells were performed. Results shown are the combined results of three separate experiments. Cell fractionation GH3 cells were seeded onto 10-cm cell culture dishes and the following day, AIP-Myc or EV (pcDNA) were transfected along with Cα-HA and R1α-HA. After 24 h, cells were counted and equal numbers were lysed and cell fractionation was performed using the subcellular protein fractionation kit (Thermo Fisher Scientific) according to manufacturer’s instructions. Lysates were subjected to Western blotting (described below). Western blotting Total lysates were prepared from cell pellets using the M-PER protein extraction reagent (Thermo Fisher Scientific) containing protease inhibitors (Complete, Roche, Basel, Switzerland). Protein concentrations were determined by BCA protein assay (Thermo Fisher Scientific) and 15 μg of lysates were subjected to SDS–PAGE and Western Blotting by standard laboratory methods. Membranes were probed with anti-AIP (Novus Biologicals) (10,33,39), anti-HA (Abcam), anti-Myc (Sigma-Aldrich), anti-GAPDH (Santa Cruz), anti-Histone H3 (Abcam), anti-TORC (Merck-Millipore, Billerica, MA), anti-HPRT (Santa Cruz), anti-e-cadherin (Abcam) or anti-α-Tubulin (Abcam), as stated, followed by appropriate HRP-coupled secondary antibodies (Jackson ImmunoResearch). Signals were detected by enhanced chemiluminescence and visualized on a ChemiDoc imaging system (Bio-Rad Laboratories, Hercules, CA). Co-immunoprecipitations For co-IPs, HEK293 cells were seeded onto 10-cm cell culture dishes and the following day, different combinations of AIP-Myc, R1α-HA, Cα-HA, untagged R1α-pcDNA or Cα-pCMV, HSP90-HA or HSC70-HA, as stated, were co-transfected. After 24 h, crosslinking was performed using 16% paraformaldehyde for 15’. Co-IPs were performed using the Pierce Co-Immunoprecipitation Kit (Thermo Fisher Scientific), in which anti-Myc or anti-HA antibodies (as stated) were covalently linked to an amine-reactive resin, according to manufacturer’s instructions. Briefly, cells were lysed, the lysate was pre-cleared using negative control beads, and some material was retained for the input control lanes, after which co-IPs with the antibody-coupled resin (10 μg antibody and 10 μl resin per reaction) were performed at 4°C overnight. Resin coupled to IgG from the same species as the IP antibody (mouse or rabbit IgG, both from Sigma-Aldrich) was used as a negative control. The following day, beads were washed three times and the bound fraction was eluted. The elution fractions were subjected to Western blotting alongside the pre-IP input lysate. PKA activity PKA activity was determined by the measurement of 32P incorporation as described previously (40). Briefly, cells were lysed and homogenized in 10 mM Tris–HCl, pH 7.0, 1 mM ethylenediaminetetraacetic acid, 0.1 mM dithiothreitol and protease inhibitor cocktail I (Merck, Kenilworth, NJ) with or without 5 µM cAMP. Samples were incubated at 37°C for 15 min after which the reaction was stopped on ice. Samples were spotted onto phosphocellulose filters, washed in phosphoric acid and dried. Analysis was performed by liquid scintillation counting. CRE-activity, double luciferase assay GH3 cells were seeded into 12-well plates and left overnight to recover. The following day, cells were transfected with a firefly luciferase reporter vector driven by CRE (CRE-Luc) and a Renilla luciferase control reporter vector (R-Luc) as well as siRNA, as stated. Firefly luciferase and Renilla luciferase activities were assessed with a dual luciferase reporter assay system (Promega) according to manufacturer’s instructions on a FLUOstar Omega luminometer (BMG Labtech, Ortenberg, Germany). PDE inhibitors used were: the nonspecific PDE inhibitor 3-isobutyl-1-methylxanthine (IBMX, 10 μM, Sigma-Aldrich), the PDE2 inhibitor erythro-9-amino-β-hexyl-α-methyl-9H-ethanol hydrochloride (EHNA, 10 μM, Sigma-Aldrich), the PDE4 inhibitor Rolipram (10 μM, Sigma-Aldrich), the PDE2A inhibitor BAY 60–7550 (50 nM, Cayman Chemical, Ann Arbor, MI) and the PDE11 inhibitor BC 11–38 (20 µM, R&D Systems, Minneapolis, MN) (41). Statistics Bar charts show means ±S.E.M. Differences between means were computed by independent samples Student’s t-tests, one-way ANOVA or two-way ANOVA followed by post hoc testing, as appropriate, using the Prism software (GraphPad, La Jolla, CA). Supplementary Material Supplementary Material is available at HMG online. Acknowledgements The project was supported by the Intramural Research Program, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health (grant number Z01-HD008920), and by the Austrian Science Fund (FWF) (Erwin Schrödinger Fellowship, grant number J3482-B13) to M.H.S.‐R. We would like to thank Prof. Márta Korbonits (Barts and the London School of Medicine and Dentistry, Queen Mary University of London) for providing the AIP, HSC70 and HSP90 plasmids, as well as Dr Forbes D. Porter and Dr Anthony Cougnoux (NICHD, NIH) for technical advice on co-immunoprecipitations. Finally, we thank Dr Vincent Schram (NICHD, NIH) and the NICHD Microscopy and Imaging Core for technical advice in co-immunofluorescence and FRET experiments and Prof. Charles Hoffman (Biology Department, Boston College) for providing the PDE11 inhibitor BC11–38. Conflict of Interest statement. None declared. References 1 Vierimaa O. , Georgitsi M. , Lehtonen R. , Vahteristo P. , Kokko A. , Raitila A. , Tuppurainen K. , Ebeling T.M. , Salmela P.I. , Paschke R. ( 2006 ) Pituitary adenoma predisposition caused by germline mutations in the AIP gene . Science , 312 , 1228 – 1230 . Google Scholar CrossRef Search ADS PubMed 2 Daly A.F. , Vanbellinghen J.F. , Khoo S.K. , Jaffrain-Rea M.L. , Naves L.A. , Guitelman M.A. , Murat A. , Emy P. , Gimenez-Roqueplo A.P. , Tamburrano G. et al. ( 2007 ) Aryl hydrocarbon receptor-interacting protein gene mutations in familial isolated pituitary adenomas: analysis in 73 families . J. Clin. Endocrinol. Metab ., 92 , 1891 – 1896 . Google Scholar CrossRef Search ADS PubMed 3 Igreja S. , Chahal H.S. , King P. , Bolger G.B. , Srirangalingam U. , Guasti L. , Chapple J.P. , Trivellin G. , Gueorguiev M. , Guegan K. et al. ( 2010 ) Characterization of aryl hydrocarbon receptor interacting protein (AIP) mutations in familial isolated pituitary adenoma families . Hum. Mutat ., 31 , 950 – 960 . Google Scholar CrossRef Search ADS PubMed 4 Hernandez-Ramirez L.C. , Gabrovska P. , Denes J. , Stals K. , Trivellin G. , Tilley D. , Ferrau F. , Evanson J. , Ellard S. , Grossman A.B. et al. ( 2015 ) Landscape of familial isolated and young-onset pituitary adenomas: prospective diagnosis in AIP mutation carriers . J. Clin. Endocrinol. Metab ., 100 , E1242 – E1254 . Google Scholar CrossRef Search ADS PubMed 5 Trivellin G. , Korbonits M. ( 2011 ) AIP and its interacting partners . J. Endocrinol ., 210 , 137 – 155 . Google Scholar CrossRef Search ADS PubMed 6 Bolger G.B. , Peden A.H. , Steele M.R. , MacKenzie C. , McEwan D.G. , Wallace D.A. , Huston E. , Baillie G.S. , Houslay M.D. ( 2003 ) Attenuation of the activity of the cAMP-specific phosphodiesterase PDE4A5 by interaction with the immunophilin XAP2 . J. Biol. Chem ., 278 , 33351 – 33363 . Google Scholar CrossRef Search ADS PubMed 7 de Oliveira S.K. , Hoffmeister M. , Gambaryan S. , Muller-Esterl W. , Guimaraes J.A. , Smolenski A.P. ( 2007 ) Phosphodiesterase 2A forms a complex with the co-chaperone XAP2 and regulates nuclear translocation of the aryl hydrocarbon receptor . J. Biol. Chem ., 282 , 13656 – 13663 . Google Scholar CrossRef Search ADS PubMed 8 Nakata A. , Urano D. , Fujii-Kuriyama Y. , Mizuno N. , Tago K. , Itoh H. ( 2009 ) G-protein signalling negatively regulates the stability of aryl hydrocarbon receptor . EMBO Rep ., 10 , 622 – 628 . Google Scholar CrossRef Search ADS PubMed 9 Hernandez-Ramirez L.C. , Trivellin G. , Stratakis C.A. ( 2017 ) Role of phosphodiesterases on the function of aryl hydrocarbon receptor-interacting protein (AIP) in the pituitary gland and on the evaluation of AIP gene variants . Horm. Metab. Res ., 49 , 286 – 295 . Google Scholar CrossRef Search ADS PubMed 10 Formosa R. , Xuereb-Anastasi A. , Vassallo J. ( 2013 ) Aip regulates cAMP signalling and GH secretion in GH3 cells . Endocr. Relat. Cancer , 20 , 495 – 505 . Google Scholar CrossRef Search ADS PubMed 11 Lania A.G. , Mantovani G. , Ferrero S. , Pellegrini C. , Bondioni S. , Peverelli E. , Braidotti P. , Locatelli M. , Zavanone M.L. , Ferrante E. et al. ( 2004 ) Proliferation of transformed somatotroph cells related to low or absent expression of protein kinase a regulatory subunit 1A protein . Cancer Res ., 64 , 9193 – 9198 . Google Scholar CrossRef Search ADS PubMed 12 Vitali E. , Peverelli E. , Giardino E. , Locatelli M. , Lasio G.B. , Beck-Peccoz P. , Spada A. , Lania A.G. , Mantovani G. ( 2014 ) Cyclic adenosine 3′-5′-monophosphate (cAMP) exerts proliferative and anti-proliferative effects in pituitary cells of different types by activating both cAMP-dependent protein kinase A (PKA) and exchange proteins directly activated by cAMP (Epac) . Mol. Cell. Endocrinol ., 383 , 193 – 202 . Google Scholar CrossRef Search ADS PubMed 13 Landis C.A. , Masters S.B. , Spada A. , Pace A.M. , Bourne H.R. , Vallar L. ( 1989 ) GTPase inhibiting mutations activate the alpha chain of Gs and stimulate adenylyl cyclase in human pituitary tumours . Nature , 340 , 692 – 696 . Google Scholar CrossRef Search ADS PubMed 14 Vallar L. , Spada A. , Giannattasio G. ( 1987 ) Altered Gs and adenylate cyclase activity in human GH-secreting pituitary adenomas . Nature , 330 , 566 – 568 . Google Scholar CrossRef Search ADS PubMed 15 Freda P.U. , Chung W.K. , Matsuoka N. , Walsh J.E. , Kanibir M.N. , Kleinman G. , Wang Y. , Bruce J.N. , Post K.D. ( 2007 ) Analysis of GNAS mutations in 60 growth hormone secreting pituitary tumors: correlation with clinical and pathological characteristics and surgical outcome based on highly sensitive GH and IGF-I criteria for remission . Pituitary , 10 , 275 – 282 . Google Scholar CrossRef Search ADS PubMed 16 Valimaki N. , Demir H. , Pitkanen E. , Kaasinen E. , Karppinen A. , Kivipelto L. , Schalin-Jantti C. , Aaltonen L.A. , Karhu A. ( 2015 ) Whole-genome sequencing of growth hormone (GH)-secreting pituitary adenomas . J. Clin. Endocrinol. Metab ., 100 , 3918 – 3927 . Google Scholar CrossRef Search ADS PubMed 17 Akintoye S.O. , Chebli C. , Booher S. , Feuillan P. , Kushner H. , Leroith D. , Cherman N. , Bianco P. , Wientroub S. , Robey P.G. et al. ( 2002 ) Characterization of gsp-mediated growth hormone excess in the context of McCune-Albright syndrome . J. Clin. Endocrinol. Metab ., 87 , 5104 – 5112 . Google Scholar CrossRef Search ADS PubMed 18 Vortmeyer A.O. , Glasker S. , Mehta G.U. , Abu-Asab M.S. , Smith J.H. , Zhuang Z. , Collins M.T. , Oldfield E.H. ( 2012 ) Somatic GNAS mutation causes widespread and diffuse pituitary disease in acromegalic patients with McCune-Albright syndrome . J. Clin. Endocrinol. Metab ., 97 , 2404 – 2413 . Google Scholar CrossRef Search ADS PubMed 19 Kirschner L.S. , Carney J.A. , Pack S.D. , Taymans S.E. , Giatzakis C. , Cho Y.S. , Cho-Chung Y.S. , Stratakis C.A. ( 2000 ) Mutations of the gene encoding the protein kinase A type I-alpha regulatory subunit in patients with the Carney complex . Nat. Genet ., 26 , 89 – 92 . Google Scholar CrossRef Search ADS PubMed 20 Robinson-White A. , Hundley T.R. , Shiferaw M. , Bertherat J. , Sandrini F. , Stratakis C.A. ( 2003 ) Protein kinase-A activity in PRKAR1A-mutant cells, and regulation of mitogen-activated protein kinases ERK1/2 . Hum. Mol. Genet ., 12 , 1475 – 1484 . Google Scholar CrossRef Search ADS PubMed 21 Skroblin P. , Grossmann S. , Schafer G. , Rosenthal W. , Klussmann E. ( 2010 ) Mechanisms of protein kinase A anchoring . Int. Rev. Cell Mol. Biol ., 283 , 235 – 330 . Google Scholar CrossRef Search ADS PubMed 22 Stojilkovic S.S. , Kretschmannova K. , Tomic M. , Stratakis C.A. ( 2012 ) Dependence of the excitability of pituitary cells on cyclic nucleotides . J. Neuroendocrinol ., 24 , 1183 – 1200 . Google Scholar CrossRef Search ADS PubMed 23 Peverelli E. , Ermetici F. , Filopanti M. , Elli F.M. , Ronchi C.L. , Mantovani G. , Ferrero S. , Bosari S. , Beck-Peccoz P. , Lania A. et al. ( 2009 ) Analysis of genetic variants of phosphodiesterase 11A in acromegalic patients . Eur. J. Endocrinol ., 161 , 687 – 694 . Google Scholar CrossRef Search ADS PubMed 24 Bolger G.B. , Bizzi M.F. , Pinheiro S.V. , Trivellin G. , Smoot L. , Accavitti M.A. , Korbonits M. , Ribeiro-Oliveira A. Jr ( 2016 ) cAMP-specific PDE4 phosphodiesterases and AIP in the pathogenesis of pituitary tumors . Endocr. Relat. Cancer , 23 , 419 – 431 . Google Scholar CrossRef Search ADS PubMed 25 Persani L. , Borgato S. , Lania A. , Filopanti M. , Mantovani G. , Conti M. , Spada A. ( 2001 ) Relevant cAMP-specific phosphodiesterase isoforms in human pituitary: effect of Gs(alpha) mutations . J. Clin. Endocrinol. Metab ., 86 , 3795 – 3800 . Google Scholar PubMed 26 Hernández-Ramírez L.C. , Morgan R.M.L. , Barry S. , D’Acquisto F. , Prodromou C. , Korbonits M. ( 2018 ) Multi-chaperone function modulation and association with cytoskeletal proteins are key features of the function of AIP in the pituitary gland . Oncotarget , 9 , 9177 – 9198 . Google Scholar CrossRef Search ADS PubMed 27 Leontiou C.A. , Gueorguiev M. , van der Spuy J. , Quinton R. , Lolli F. , Hassan S. , Chahal H.S. , Igreja S.C. , Jordan S. , Rowe J. et al. ( 2008 ) The role of the aryl hydrocarbon receptor-interacting protein gene in familial and sporadic pituitary adenomas . J. Clin. Endocrinol. Metab ., 93 , 2390 – 2401 . Google Scholar CrossRef Search ADS PubMed 28 Morgan R.M. , Hernandez-Ramirez L.C. , Trivellin G. , Zhou L. , Roe S.M. , Korbonits M. , Prodromou C. ( 2012 ) Structure of the TPR domain of AIP: lack of client protein interaction with the C-terminal alpha-7 helix of the TPR domain of AIP is sufficient for pituitary adenoma predisposition . PLoS One , 7 , e53339. Google Scholar CrossRef Search ADS PubMed 29 Hernandez-Ramirez L.C. , Martucci F. , Morgan R.M. , Trivellin G. , Tilley D. , Ramos-Guajardo N. , Iacovazzo D. , D’Acquisto F. , Prodromou C. , Korbonits M. ( 2016 ) Rapid proteasomal degradation of mutant proteins is the primary mechanism leading to tumorigenesis in patients with missense AIP mutations . J. Clin. Endocrinol. Metab ., 101 , 3144 – 3154 . Google Scholar CrossRef Search ADS PubMed 30 Tuominen I. , Heliovaara E. , Raitila A. , Rautiainen M.R. , Mehine M. , Katainen R. , Donner I. , Aittomaki V. , Lehtonen H.J. , Ahlsten M. et al. ( 2015 ) AIP inactivation leads to pituitary tumorigenesis through defective Galphai-cAMP signaling . Oncogene , 34 , 1174 – 1184 . Google Scholar CrossRef Search ADS PubMed 31 Oesch-Bartlomowicz B. , Oesch F. ( 2009 ) Role of cAMP in mediating AHR signaling . Biochem. Pharmacol ., 77 , 627 – 641 . Google Scholar CrossRef Search ADS PubMed 32 Oesch-Bartlomowicz B. , Huelster A. , Wiss O. , Antoniou-Lipfert P. , Dietrich C. , Arand M. , Weiss C. , Bockamp E. , Oesch F. ( 2005 ) Aryl hydrocarbon receptor activation by cAMP vs. dioxin: divergent signaling pathways . Proc. Natl. Acad. Sci. U. S. A ., 102 , 9218 – 9223 . Google Scholar CrossRef Search ADS PubMed 33 Heliovaara E. , Raitila A. , Launonen V. , Paetau A. , Arola J. , Lehtonen H. , Sane T. , Weil R.J. , Vierimaa O. , Salmela P. et al. ( 2009 ) The expression of AIP-related molecules in elucidation of cellular pathways in pituitary adenomas . Am. J. Pathol ., 175 , 2501 – 2507 . Google Scholar CrossRef Search ADS PubMed 34 Raitila A. , Lehtonen H.J. , Arola J. , Heliovaara E. , Ahlsten M. , Georgitsi M. , Jalanko A. , Paetau A. , Aaltonen L.A. , Karhu A. ( 2010 ) Mice with inactivation of aryl hydrocarbon receptor-interacting protein (Aip) display complete penetrance of pituitary adenomas with aberrant ARNT expression . Am. J. Pathol ., 177 , 1969 – 1976 . Google Scholar CrossRef Search ADS PubMed 35 Jaffrain-Rea M.L. , Angelini M. , Gargano D. , Tichomirowa M.A. , Daly A.F. , Vanbellinghen J.F. , D’Innocenzo E. , Barlier A. , Giangaspero F. , Esposito V. et al. ( 2009 ) Expression of aryl hydrocarbon receptor (AHR) and AHR-interacting protein in pituitary adenomas: pathological and clinical implications . Endocr. Relat. Cancer , 16 , 1029 – 1043 . Google Scholar CrossRef Search ADS PubMed 36 Lecoq A.L. , Viengchareun S. , Hage M. , Bouligand J. , Young J. , Boutron A. , Zizzari P. , Lombes M. , Chanson P. , Kamenicky P. ( 2016 ) AIP mutations impair AhR signaling in pituitary adenoma patients fibroblasts and in GH3 cells . Endocr. Relat. Cancer , 23 , 433 – 443 . Google Scholar CrossRef Search ADS PubMed 37 Kirschner L.S. , Yin Z. , Jones G.N. , Mahoney E. ( 2009 ) Mouse models of altered protein kinase A signaling . Endocr. Relat. Cancer , 16 , 773 – 793 . Google Scholar CrossRef Search ADS PubMed 38 Mavrakis M. , Lippincott-Schwartz J. , Stratakis C.A. , Bossis I. ( 2006 ) Depletion of type IA regulatory subunit (RIalpha) of protein kinase A (PKA) in mammalian cells and tissues activates mTOR and causes autophagic deficiency . Hum. Mol. Genet ., 15 , 2962 – 2971 . Google Scholar CrossRef Search ADS PubMed 39 Chahal H.S. , Trivellin G. , Leontiou C.A. , Alband N. , Fowkes R.C. , Tahir A. , Igreja S.C. , Chapple J.P. , Jordan S. , Lupp A. et al. ( 2012 ) Somatostatin analogs modulate AIP in somatotroph adenomas: the role of the ZAC1 pathway . J. Clin. Endocrinol. Metab ., 97 , E1411 – E1420 . Google Scholar CrossRef Search ADS PubMed 40 Nesterova M.V. , Sashchenko L.P. , Vasiliev V.Y. , Severin E.S. ( 1975 ) A cyclic adenosine 3′, 5′-monophosphate-dependent histone kinase from pig brain. Purification and some properties of the enzyme . Biochim. Biophys. Acta , 377 , 271 – 281 . Google Scholar CrossRef Search ADS PubMed 41 Ceyhan O. , Birsoy K. , Hoffman C.S. ( 2012 ) Identification of biologically active PDE11-selective inhibitors using a yeast-based high-throughput screen . Chem. Biol ., 19 , 155 – 163 . Google Scholar CrossRef Search ADS PubMed Published by Oxford University Press 2018. This work is written by US Government employees and is in the public domain in the US. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Human Molecular Genetics Oxford University Press

Interaction of AIP with protein kinase A (cAMP-dependent protein kinase)

Loading next page...
 
/lp/ou_press/interaction-of-aip-with-protein-kinase-a-camp-dependent-protein-kinase-np1n1C4p51
Publisher
Oxford University Press
Copyright
Published by Oxford University Press 2018. This work is written by US Government employees and is in the public domain in the US.
ISSN
0964-6906
eISSN
1460-2083
D.O.I.
10.1093/hmg/ddy166
Publisher site
See Article on Publisher Site

Abstract

Abstract Germline mutations in the aryl hydrocarbon receptor-interacting protein (AIP) gene cause mostly somatotropinomas and/or prolactinomas in a subset of familial isolated pituitary adenomas (FIPA). AIP has been shown to interact with phosphodiesterases (PDEs) and G proteins, suggesting a link to the cyclic AMP (cAMP)-dependent protein kinase (PKA) pathway. Upregulation of PKA is seen in sporadic somatotropinomas that carry GNAS mutations, and those in Carney complex that are due to PRKAR1A mutations. To elucidate the mechanism of AIP-dependent pituitary tumorigenesis, we studied potential functional and physical interactions of AIP with PKA’s main subunits PRKAR1A (R1α) and PRKACA (Cα). We found that AIP physically interacts with both R1α and Cα; this interaction is enhanced when all three components are present, but maintained during Cα-R1α dissociation by PKA activation, indicating that AIP binds Cα/R1α both in complex and separately. The interaction between AIP and R1α/Cα is reduced when the frequent AIP pathogenic mutation p.R304* is present. AIP protein levels are regulated both by translation and the ubiquitin/proteasome pathway and Cα stabilizes both AIP and R1α protein levels. AIP reduction by siRNA leads to an increase of PKA activity, which is disproportionately enhanced during PDE4-inhibition. We show that AIP interacts with the PKA pathway on multiple levels, including a physical interaction with both the main regulatory (R1α) and catalytic (Cα) PKA subunits and a functional interaction with PDE4-dependent PKA activation. These findings provide novel insights on the mechanisms of AIP-dependent pituitary tumorigenesis. Introduction Germline mutations in the aryl hydrocarbon receptor interacting protein (AIP, OMIM*605555) have recently been identified as the genetic cause of ∼20% of familial isolated pituitary adenomas (FIPA, OMIM#102200) (1–3). AIP-associated pituitary adenomas are mostly early-onset, aggressive somatotropinomas, while patients demonstrate no other recurrent clinical manifestations (2–4). The molecular mechanism by which these inactivating AIP mutations lead to pituitary tumor formation is currently unclear. AIP interacts with components of the cyclic adenosine monophosphate (cAMP)/protein kinase A (PKA) pathway, including the phosphodiesterases (PDEs) 4A5 and 2A3, as well as the G proteins Gα13 and Gαq (5–9). Functionally, a reduction of Aip levels in GH3 rat pituitary somatotroph cells led to an increase in PKA signaling on different levels (10); however, this effect was not completely abolished by PDE inhibition (10), suggesting that other components of the PKA pathway mediate AIP-dependent effects on pituitary tumor development. The cAMP/PKA pathway has been shown to be involved in the regulation of pituitary cell proliferation and pituitary hormone secretion and it is frequently dysregulated in pituitary tumors (11,12). Approximately 40% of sporadic somatotropinomas were shown to carry somatic mutations of the α-subunit of the Gs protein (GNAS), leading to adenylyl cyclase and hence PKA pathway activation (13–16). McCune-Albright syndrome (OMIM#174800) is caused by a postzygotic mutation of GNAS and ∼20% of patients present with somatotroph or somato-lactotroph hyperplasia and growth hormone (GH) hypersecretion (17,18). Carney complex (CNC, OMIM#160980)-associated somatotropinomas are caused by inactivating mutations of the PKA regulatory subunit 1α (PRKAR1A), leading to a lack of inhibition of the PKA catalytic subunits, mainly Cα (19,20). Even in the absence of PRKAR1A mutations or loss of heterozygosity, R1α protein expression is markedly reduced in sporadic pituitary tumors including somatotropinomas, which contributes to somatotroph proliferation (11). In the present investigation, we examined the hypothesis that AIP may functionally and physically interact with components of the PKA pathway, including the main PKA regulatory (R1α) and catalytic (Cα) subunit, encoded, respectively, by the PRKAR1A and PRKACA genes. The data uncover new AIP interactions that point to cAMP/PKA’s involvement in the former’s tumor suppressor role. Results Intracellular localization and co-localization of AIP, R1α and Cα Confocal immunofluorescence shows that endogenous AIP colocalizes with R1α and Cα in the rat mammosomatotropinoma cell line GH3 (Fig. 1A and B). Colocalization was observed mainly in cytoplasmic foci, potentially relating to the cytoplasmic anchoring of R1α to A-kinase anchoring proteins (AKAPs) (21). The colocalization between AIP and Cα significantly increases during PKA activation by forskolin (fsk) compared to vehicle (Pearson’s R above threshold 0.765 versus 0.721, P = 0.012). Fsk, which is widely used to stimulate pituitary hormone secretion in vitro, activates adenylyl cyclase, raises intracellular cAMP levels and thereby activates PKA (10,22). Colocalization decreases in the presence of the PKA inhibitor H-89 (Pearson’s R above threshold 0.693, P = 0.006, Supplementary Material, Fig. S1). AIP colocalization with R1α is not significantly affected by fsk or H-89 (Supplementary Material, Fig. S1). Figure 1. View largeDownload slide Intracellular localization and colocalization of AIP, Cα and R1α. (A and B) Confocal immunofluorescence of AIP (green channel) with Cα (red channel) (A) or with R1α (B) in GH3 cells at 63× magnification. Nuclei were visualized with DAPI (blue channel). Cells were treated with 10 µM forskolin (fsk) or DMSO (vehicle) for 30 min. (C) GH3 cells overexpressing AIP (AIP +) or empty vector (AIP−) were subjected to cell fractionation to separate cytosolic (Cyt), membranous (mem), nuclear and nuclear bound (N. b.) compartments. Fractions were subjected to Western blotting and R1ɑ, Cα and AIP as well as GAPDH, GR, e-cadherin, histone H3, HPRT and TORC were detected. Figure 1. View largeDownload slide Intracellular localization and colocalization of AIP, Cα and R1α. (A and B) Confocal immunofluorescence of AIP (green channel) with Cα (red channel) (A) or with R1α (B) in GH3 cells at 63× magnification. Nuclei were visualized with DAPI (blue channel). Cells were treated with 10 µM forskolin (fsk) or DMSO (vehicle) for 30 min. (C) GH3 cells overexpressing AIP (AIP +) or empty vector (AIP−) were subjected to cell fractionation to separate cytosolic (Cyt), membranous (mem), nuclear and nuclear bound (N. b.) compartments. Fractions were subjected to Western blotting and R1ɑ, Cα and AIP as well as GAPDH, GR, e-cadherin, histone H3, HPRT and TORC were detected. In cell fractionation experiments, AIP is primarily localized within the cytoplasm and in the membranous fraction (Fig. 1C). Similarly, R1α and Cα mainly localize to the cytoplasm and membranes; a small amount of Cα is detectable in the unbound nuclear fraction. The purity of fractions and equal loading were verified by visualizing the expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH, expressed in the cytosol and perimembranous), the glucocorticoid receptor (GR, expressed in the cell membrane and in the nucleus), e-cadherin (cell membrane expression), histone H3 (chromatin-bound in the nuclear bound fraction), hypoxanthine-guanine phosphoribosyltransferase (HPRT, cytosolic expression) and the transcriptional co-activator transducer of regulated CREB-binding proteins (TORC, chromatin-bound). AIP overexpression leads to a decrease in membranous R1α levels as well as a decrease in nuclear Cα expression, compared to control (empty vector). The specificity of the AIP antibody was verified during AIP overexpression in a separate experiment (Supplementary Material, Fig. S2). Physical interaction between AIP, R1α and Cα The influence of AIP on the interaction between R1α and Cα was tested in fluorescence resonance energy transfer (FRET) by acceptor photobleaching. R1α-Venus and Cα-Cerulean were co-overexpressed with or without AIP, and their interaction was measured by the increase of Cerulean fluorescence intensity after photobleaching Venus. Forskolin treatment caused a decrease in donor intensity to 42% (P < 0.001) relative to control, consistent with dissociation and decreased interaction between R1α and Cα after PKA pathway activation (Fig. 2A). AIP overexpression showed a tendency towards a decrease in R1α-Cα interaction in basal conditions, although this did not reach statistical significance. In a two-way ANOVA an interaction effect was detected, suggesting a synergistic effect of Fsk and AIP overexpression in the dissociation of the PKA subunits (P = 0.002 for interaction effects between Fsk*AIP overexpression in two-way ANOVA). Figure 2. View largeDownload slide Physical interaction between AIP and the PKA. (A) GH3 cells were co-transfected with R1α-Venus and Cα-Cerulean as well as either empty vector (EV) or AIP. After 24 h, cells were stimulated with 10 µM forskolin (Fsk) or vehicle for 1 h before imaging. The interaction between R1α and Cα was then quantified via fluorescence resonance energy transfer (FRET) by acceptor (R1α-Venus) photobleaching. Differences between means were tested by two-way ANOVA followed by post hoc testing. ***P < 0.001, n.s. not significant. Data are expressed relative to EV/vehicle control and represent the mean of three separate experiments. (B) Co-IP of AIP-Myc on Myc-coupled resin with R1α and HSC70. AIP-Myc, HSC70-HA and R1α-HA were overexpressed; IPs were performed with or without crosslinking with paraformaldehyde (PFA). (C) Co-IP of AIP-Myc on Myc-coupled resin with R1α and HSP90. AIP-Myc, HSP90-HA and R1α-HA were overexpressed; IPs were performed with or without crosslinking with paraformaldehyde (PFA). (D) Co-IP of R1α-HA on HA-coupled resin with AIP-Myc. R1α-HA and AIP-Myc were overexpressed alone or in combination (top panel). IgG alone was run in the left lane of each Western blot, while the right two lanes show the respective controls with IgG-coupled resin. (E) Co-IP of AIP-Myc on Myc-coupled resin with R1α-HA and Cα-HA. Different combinations of R1α-HA, Cα-HA and AIP-Myc were overexpressed (top panel). The left two lanes were additionally stimulated with 10 µM forskolin (Fsk). (F) Co-IP of Cα-HA on HA-coupled resin with AIP-Myc and R1α. Cα-HA, R1α-pcDNA and either AIP-Myc wild-type (AIP wt) or AIP-Myc R304* (AIP R304*) were overexpressed (top panel). (G) Co-IP of R1α-HA on HA-coupled resin with AIP-Myc and Cα. R1α-HA, Cα-pCMV and either AIP-Myc wild-type (AIP wt) or AIP-Myc R30`4* (AIP R304*) were overexpressed (top panel). Figure 2. View largeDownload slide Physical interaction between AIP and the PKA. (A) GH3 cells were co-transfected with R1α-Venus and Cα-Cerulean as well as either empty vector (EV) or AIP. After 24 h, cells were stimulated with 10 µM forskolin (Fsk) or vehicle for 1 h before imaging. The interaction between R1α and Cα was then quantified via fluorescence resonance energy transfer (FRET) by acceptor (R1α-Venus) photobleaching. Differences between means were tested by two-way ANOVA followed by post hoc testing. ***P < 0.001, n.s. not significant. Data are expressed relative to EV/vehicle control and represent the mean of three separate experiments. (B) Co-IP of AIP-Myc on Myc-coupled resin with R1α and HSC70. AIP-Myc, HSC70-HA and R1α-HA were overexpressed; IPs were performed with or without crosslinking with paraformaldehyde (PFA). (C) Co-IP of AIP-Myc on Myc-coupled resin with R1α and HSP90. AIP-Myc, HSP90-HA and R1α-HA were overexpressed; IPs were performed with or without crosslinking with paraformaldehyde (PFA). (D) Co-IP of R1α-HA on HA-coupled resin with AIP-Myc. R1α-HA and AIP-Myc were overexpressed alone or in combination (top panel). IgG alone was run in the left lane of each Western blot, while the right two lanes show the respective controls with IgG-coupled resin. (E) Co-IP of AIP-Myc on Myc-coupled resin with R1α-HA and Cα-HA. Different combinations of R1α-HA, Cα-HA and AIP-Myc were overexpressed (top panel). The left two lanes were additionally stimulated with 10 µM forskolin (Fsk). (F) Co-IP of Cα-HA on HA-coupled resin with AIP-Myc and R1α. Cα-HA, R1α-pcDNA and either AIP-Myc wild-type (AIP wt) or AIP-Myc R304* (AIP R304*) were overexpressed (top panel). (G) Co-IP of R1α-HA on HA-coupled resin with AIP-Myc and Cα. R1α-HA, Cα-pCMV and either AIP-Myc wild-type (AIP wt) or AIP-Myc R30`4* (AIP R304*) were overexpressed (top panel). Co-immunoprecipitation (co-IP) experiments were conducted to further clarify a potential interaction of AIP with R1α or Cα. Since AIP is bound and stabilized by the two chaperones HSP90 (heat shock protein 90) and HSC70 (heat shock cognate 70), which assist in its folding and may facilitate protein–protein interactions (5), we suspected that any AIP-PKA interaction may also be stabilized by the presence of HSP90 or HSC70. We found that AIP interacts with R1α both in the presence (Fig. 2B and C) and absence (Fig. 2D) of HSC70 and HSP90, thus indicating that the AIP-R1α interaction is independent of the presence of these two chaperone proteins. In fact, the presence of HSP90 does not appear to increase the amount of R1α bound to AIP (data not shown). AIP also interacts with Cα: this interaction is increased in the presence of R1α, suggesting that an AIP-R1α-Cα complex may be a more stable configuration. Furthermore, this interaction is maintained in the presence of the PKA activator forskolin (Fig. 2E), which leads to the dissociation of the subunits, thereby suggesting that AIP can bind R1α and Cα in their inactive complex as well as the active separate subunits. Conversely, the influence of Cα on the AIP-R1α association cannot be determined since AIP and R1α protein expression levels are also highly dependent on the presence of Cα (Fig. 4). The interaction between AIP and Cα (Fig. 2F) as well as that of AIP and R1α (Fig. 2G) is reduced in the truncated AIP R304* mutant that lacks most of the C-terminal α-7 helix that is crucial for protein–protein interactions. Levels of AIP co-IP remain reduced even when correcting for the lower expression of AIP R304* compared to wild-type (R1α-AIP R304* 0.5-fold, Cα-AIP R304* 0.33-fold compared to wild-type), suggesting that the α-7 helix of AIP is involved in the interaction between AIP and R1α/Cα, but other parts of AIP also contribute to this interaction. AIP reduces PKA activity Basal and total PKA activity (in the absence or presence of cAMP, respectively) were determined in GH3 cells during AIP overexpression. While no significant difference was detected in basal PKA activity between AIP and control (Fig. 3A), total PKA activity was significantly lower during AIP overexpression (Fig. 3B). We did not observe a significant effect of AIP silencing by siRNA on basal or total PKA activity (data not shown). Figure 3. View largeDownload slide PKA activity in response to AIP overexpression. AIP or empty vector (EV) were overexpressed in GH3 cells for 48 h; cells were lysed and PKA activity was determined in the absence [basal PKA activity (A)] and in the presence [total PKA activity (B)] of cAMP. Differences between means were tested by independent samples Student’s t-tests. ***P < 0.001, n.s. not significant. Data shown are expressed relative to EV and represent the mean from two separate experiments. Figure 3. View largeDownload slide PKA activity in response to AIP overexpression. AIP or empty vector (EV) were overexpressed in GH3 cells for 48 h; cells were lysed and PKA activity was determined in the absence [basal PKA activity (A)] and in the presence [total PKA activity (B)] of cAMP. Differences between means were tested by independent samples Student’s t-tests. ***P < 0.001, n.s. not significant. Data shown are expressed relative to EV and represent the mean from two separate experiments. Cα stabilizes AIP and R1α protein levels AIP protein levels are increased in the presence of Cα (Fig. 4A and D). During overexpression, AIP protein levels are ∼2-fold increased when Cα is co-overexpressed. The presence of Cα also increases R1α levels by >10-fold (Fig. 4A and D). The stabilizing effect of Cα on AIP protein levels is lightly dampened by the additional presence of R1α, which may limit Cα binding capacities to AIP. Aip mRNA levels do not significantly differ in the presence of Cα (Fig. 4B), indicating that Cα stabilizes AIP and R1α only at the protein level. Aip mRNA levels are similarly not affected during activation of the PKA pathway with fsk and consequent dissociation of the PKA subunits or during PKA pathway inhibition (when PKA subunits are in complex) with the PKA inhibitor peptide PKI (Fig. 4C). These results suggest that the stabilizing effect of Cα on AIP is independent of PKA pathway activity. Figure 4. View largeDownload slide Regulation of AIP mRNA and protein levels by Cα and R1α. (A) GH3 cells were co-transfected with different combinations of AIP, Cα and R1α (as indicated, top panel). Total cell lysate was subjected to Western blotting. Expression levels were compared to tubulin as a loading control. (B) GH3 cells were co-transfected with Cα and/or R1α, as indicated. Expression levels of Aip mRNA were determined by RT-qPCR in relation to β-Actin levels. (C) GH3 cells were treated with 10 µM forskolin (Fsk) or 1 µM PKA inhibitor (PKI) for 8 h. Aip expression levels were determined by RT-qPCR in relation to β-Actin levels. (D) GH3 cells were co-transfected with different combinations of AIP, Cα and R1α (as indicated, top panel), followed by treatment with 5 µg/ml cycloheximide (CHX), 1 µM MG132 or vehicle. Total cell lysate was subjected to Western blotting. Expression levels were compared to tubulin as a loading control. Differences between means were tested by one-way ANOVA followed by post hoc testing (B and C). n.s. not significant. Data are expressed relative to EV (B) or vehicle (C) controls and represent the mean from one representative experiment performed in triplicate wells. Figure 4. View largeDownload slide Regulation of AIP mRNA and protein levels by Cα and R1α. (A) GH3 cells were co-transfected with different combinations of AIP, Cα and R1α (as indicated, top panel). Total cell lysate was subjected to Western blotting. Expression levels were compared to tubulin as a loading control. (B) GH3 cells were co-transfected with Cα and/or R1α, as indicated. Expression levels of Aip mRNA were determined by RT-qPCR in relation to β-Actin levels. (C) GH3 cells were treated with 10 µM forskolin (Fsk) or 1 µM PKA inhibitor (PKI) for 8 h. Aip expression levels were determined by RT-qPCR in relation to β-Actin levels. (D) GH3 cells were co-transfected with different combinations of AIP, Cα and R1α (as indicated, top panel), followed by treatment with 5 µg/ml cycloheximide (CHX), 1 µM MG132 or vehicle. Total cell lysate was subjected to Western blotting. Expression levels were compared to tubulin as a loading control. Differences between means were tested by one-way ANOVA followed by post hoc testing (B and C). n.s. not significant. Data are expressed relative to EV (B) or vehicle (C) controls and represent the mean from one representative experiment performed in triplicate wells. To investigate the cellular mechanisms governing AIP, R1α and Cα protein levels, experiments were conducted using the inhibitor of protein synthesis cycloheximide (CHX) and the proteasome inhibitor MG132. In the absence of R1α and Cα overexpression, CHX leads to a reduction of overexpressed AIP levels within 14 h, while the addition of MG132 rescues this effect, indicating that AIP protein levels are regulated by translation and degradation via the ubiquitin/proteasome pathway (Fig. 4D). R1α levels are increased after proteasome inhibition, whereas Cα levels are more strongly influenced by translational inhibition. When all three AIP, R1α and Cα are co-overexpressed there appears to be a stronger overall effect of CHX than MG132, indicating that the three proteins together are strongly regulated by translation (Fig. 4D). AIP functionally interacts with PDE-dependent PKA pathway activity via PDE4 AIP is known to directly interact with PDE2 and PDE4 (5), while genetic variants of PDE11A were recently investigated in the context of acromegaly (23). We first tested the effect of different subtype-specific PDE inhibitors on cAMP-responsive element (CRE)-activation in a double luciferase assay in GH3 cells. The PDE2 inhibitors EHNA and BAY 60–7550, and the PDE11A inhibitor BC11–38 did not have an effect on CRE-activity, while the pan-PDE inhibitor IBMX and the PDE4 inhibitor Rolipram increased basal CRE-activity (Fig. 5A), suggesting a physiologically important role of PDE4 in PKA signaling in pituitary somatotrophs (24,25). Figure 5. View largeDownload slide Functional interaction between AIP and PDE-dependent PKA pathway activity. (A) GH3 cells were co-transfected with a cAMP responsive element luciferase reporter vector (CRE-Luc) and a control Renilla luciferase vector (R-Luc). The following day, cells were stimulated with different PDE inhibitors or vehicle, as stated, for 8 h, after which CRE-Luc/R-Luc were assessed in a double luciferase assay. (B) GH3 cells were co-transfected with CRE-Luc, R-Luc and Aip siRNA or control siRNA. The following day, cells were stimulated with either IBMX, Rolipram or vehicle for 8 h, followed by a double luciferase assay. Differences between means were tested by one-way ANOVA (A) or two-way ANOVA (B) followed by post hoc testing. *P < 0.05, ***P < 0.001, n.s. not significant. Data shown are expressed relative to vehicle (A) or vehicle/EV (B) controls and represent the mean from one representative experiment performed in triplicate wells. Figure 5. View largeDownload slide Functional interaction between AIP and PDE-dependent PKA pathway activity. (A) GH3 cells were co-transfected with a cAMP responsive element luciferase reporter vector (CRE-Luc) and a control Renilla luciferase vector (R-Luc). The following day, cells were stimulated with different PDE inhibitors or vehicle, as stated, for 8 h, after which CRE-Luc/R-Luc were assessed in a double luciferase assay. (B) GH3 cells were co-transfected with CRE-Luc, R-Luc and Aip siRNA or control siRNA. The following day, cells were stimulated with either IBMX, Rolipram or vehicle for 8 h, followed by a double luciferase assay. Differences between means were tested by one-way ANOVA (A) or two-way ANOVA (B) followed by post hoc testing. *P < 0.05, ***P < 0.001, n.s. not significant. Data shown are expressed relative to vehicle (A) or vehicle/EV (B) controls and represent the mean from one representative experiment performed in triplicate wells. To test a possible functional interaction between AIP and PDEs in GH3 cells, Aip reduction by siRNA-mediated knockdown and PDE inhibition were performed concurrently. Aip knockdown led to a modest increase of PKA activity as measured by CRE-activity (Fig. 5B). IBMX and Rolipram also lead to increased CRE-activation, which is even further increased in the absence of Aip. Levels of CRE-activation were compared during Aip siRNA, PDE inhibition by IBMX and Rolipram, and both simultaneous treatments in a two-way ANOVA. We found a significant interaction effect between Aip knockdown and PDE inhibition (P < 0.001 for interaction effects between Aip siRNA*IBMX and P < 0.001 for interaction effects between Aip siRNA*Rolipram); the effect of Aip siRNA on CRE-activation is proportionately higher during PDE-inhibition by IBMX and Rolipram compared to control [1.45-fold (control) versus 2.37-fold increase for IBMX and 1.75-fold increase for Rolipram]. This suggests that AIP functionally interacts with PDE activity in pituitary somatotrophs, and this interaction is at least partially mediated via PDE4. Discussion We show here that there is a cytoplasmic and perimembranous colocalization of AIP with the regulatory and catalytic subunits of the PKA, whereby AIP-Cα colocalization increases during PKA pathway activation. AIP physically interacts with both R1α and Cα. This interaction is enhanced when all three components are present, but maintained during Cα-R1α dissociation by PKA activation, indicating that AIP binds Cα and R1α both in complex and separately. AIP overexpression leads to a decrease of PKA activity. The interaction between AIP and R1α/Cα is reduced in the frequent pathogenic mutant AIP p.R304*, which lacks its C-terminal α-helix. AIP protein levels are regulated by translation and the ubiquitin/proteasome pathway and Cα stabilizes both AIP and R1α protein levels. Aip reduction by siRNA leads to an increase of PKA pathway activity, which is disproportionately enhanced during PDE4-inhibition, indicating that Aip functionally interacts with PDE4-dependent PKA pathway activity. While it is known that AIP deficiency leads to altered PKA signaling in pituitary somatotrophs (10), a physical interaction between AIP and the PKA was not previously demonstrated. We show here that AIP interacts with R1α as well as Cα. Conversely, the catalytic PKA subunit stabilizes AIP protein levels. While we cannot exclude that these interactions are mediated via yet another mutual interaction partner (26), we provide evidence that both these interactions are maintained during PKA activation when both PKA subunits are separate. AIP-Cα binding may influence Cα function when AIP is bound to Cα. We show that Aip-Cα colocalization is increased during PKA activation and the dissociation of regulatory and catalytic PKA subunits. The observed cytoplasmic colocalization between AIP and R1α as well as Cα was not uniform but confined to some cytoplasmic areas. The intracellular localization of the PKA is known to be regulated via the AKAPs which act as a scaffold for the PKA regulatory and catalytic subunits as well as PKA substrates and other components of the PKA pathway, thereby regulating PKA pathway activity and specificity (21). AIP was also previously shown to localize to secretory granules in somatotrophs (27). Furthermore, we show that AIP overexpression altered nuclear Cα levels, suggesting that AIP-Cα binding may modify Cα localization and hence its activity; this could be another mechanism by which AIP deficiency may contribute to pituitary tumor formation. Functionally, AIP overexpression led to a decrease in total PKA activity, and additionally showed a trend towards lowering basal PKA activity. These findings are in agreement with the observation that AIP binds to the PKA subunits in the inactive as well as in the active configuration. Mechanistically, AIP may bind Cα and thereby inhibit its catalytic activity, or it may bind the inactive Cα-R1α complex and prevent subunit dissociation, thereby decreasing levels of active Cα. We and others (10) found that Aip silencing leads to an increase in PKA pathway activation as measured by CRE-activity (Fig. 5B), while basal and total PKA activity were not significantly changed, overall suggesting that Aip silencing causes a modest increase in PKA pathway activity that is not due to direct activation of the PKA. The truncating AIP p.R304* mutant, one of the most common mutations detected in human AIP-dependent FIPA patients, interacts with the PKA subunits to a lesser degree than the wild-type, suggesting that the binding between the PKA subunits and AIP may contribute to the pathogenic mechanism of human AIP-dependent pituitary tumorigenesis. Furthermore, these results demonstrate that the C-terminal α-7 helix of AIP, which is truncated in AIP p.R304*, is involved in but not solely responsible for AIP-R1α and AIP-Cα binding. To further elucidate the importance of different regions of AIP in the interaction with the PKA, and to infer the contribution of this interaction to pituitary tumorigenesis, other pathogenic mutants should be investigated in the future. AIP contains three tetratricopeptide repeat (TPR) domains, which were shown to contribute to AIP homodimerization as well as to AIP binding to a plethora of interacting partners including PDE4A5 and HSP90 (5,28). In addition, many pathogenic AIP missense mutations are localized within the TPR domains, thereby highlighting their importance in AIP-dependent pituitary tumorigenesis (5). These TPR domains could equally contribute to the binding of AIP to R1α/Cα. While AIP binds both R1α and Cα, the binding affinities to each PKA subunit is likely to be different and may contribute to potential different functional effects of the AIP-R1α and the AIP-Cα interaction. Considering this, intracellular AIP levels and hence the proportion of AIP bound to Cα or R1α or the Cα-R1α complex may be crucial in the contribution of AIP to PKA pathway activity. The reduction of half-life of AIP mutants was recently shown to be a major contributor to the pathogenic mechanism in pituitary tumorigenesis (29). We confirm and extend these findings by demonstrating that 1. AIP protein levels but not mRNA levels are strongly influenced by Cα, and to a lesser degree, R1α; 2. both Cα and R1α physically bind to AIP; 3. this binding is reduced in the AIP p.R304* mutant, which was also shown to have a severely reduced half-life (29). We propose that the alterations in AIP half-life may at least partially be mediated by impaired AIP-Cα/R1α binding, and that a reduction of this binding contributes to pituitary tumorigenesis by modifying PKA pathway activity. Our data also confirm a role of the ubiquitin proteasome pathway in the regulation of intracellular AIP levels, in addition to translational control. Either or both of those mechanisms may be responsible for the stabilizing effect of Cα (and R1α to a smaller extent) on cellular AIP levels. The intracellular action of AIP in the context of pituitary tumor formation is complex and other mechanisms are also likely to contribute to AIP-dependent pituitary tumorigenesis. Others have shown that AIP interacts with the inhibitory pathway involving Gαi (30). AIP is known to bind to PDEs (5); AIP binds and inhibits PDE4A5 catalytic activity via its TPR domain (6,9). Our data demonstrate that reduced AIP levels disproportionately enhance PKA pathway activity under PDE4-specific inhibition in pituitary somatotrophs; hence disturbance of the AIP-PDE4 interaction may contribute to AIP-dependent pituitary tumorigenesis. AIP is a member of the aryl hydrocarbon receptor (AHR) signaling pathway. The AHR binds environmental toxins and mediates the cellular transcriptional response to those toxins; however the AHR pathway is also involved in physiologically essential ligand-independent functions, including the regulation of cell growth and differentiation (31). Nuclear translocation of AHR is stimulated by cAMP (32), demonstrating an overlap between the AHR and the PKA pathways, which may form the basis of AHR’s ligand-independent functions. Since AIP also acts on cAMP levels (10), AIP could potentially be an upstream regulator of ligand-independent AHR effects in addition to PKA pathway activity; further studies are necessary to determine if AIP can indirectly affect AHR translocation via altered cAMP levels. The interaction between AIP and the PKA we show here present an additional point of overlap between the AHR and PKA pathways. In a pituitary-specific context, decreased expression of AHR and AHR nuclear translocator (ARNT) were described in AIP-dependent pituitary adenomas (33–35). Furthermore, recent data also demonstrated that transcriptional targets of the AHR pathway were altered in fibroblasts of patients carrying inactivating AIP mutations (36). R1α and Cα are regarded as the main PKA subunits; they are the most highly expressed and best studied regulatory and catalytic PKA subunits (37). A low ratio between R1α and the R2 subunits (R2α and R2β) by regulation of R1α levels in human somatotropinomas was suggested as a potential mechanism contributing to pituitary tumorigenesis (11). The role of different PKA catalytic subunits in the pituitary has not been investigated in detail; in addition to the interaction between R1α and Cα with AIP shown here, other PKA subunits may also play a role in AIP-dependent PKA pathway modulation and this should be investigated in the future. We propose that the mechanisms of AIP-dependent pituitary tumor formation are complex and involve an impact on PKA pathway activation upstream of the PKA, leading to quantitative changes of PKA activity, as well as a physical interaction with PKA subunits governing downstream effects. While this study focused primarily on the mechanisms of AIP towards pituitary tumorigenesis, AIP is an ubiquitously expressed protein whose function(s) in other tissues is still not well defined. Since both the PKA pathway and the AHR pathway are ubiquitous regulators of metabolism and proliferation, an interaction between AIP and the PKA pathway is also highly relevant for other areas of research and tumor entities. We identified here a physical interaction between AIP and the PKA pathway and its subunits, which may play an important role for pituitary tumorigenesis. A clear understanding of the molecular mechanisms of tumorigenesis is crucial for the development of novel rational therapeutic strategies. Materials and Methods Cell culture, plasmids and transfection The rat mammosomatotropinoma cell line GH3 and the human embryonic kidney cell line HEK293 were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and cells were maintained under standard cell culture conditions in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/ml penicillin, 100 μg/ml streptomycin and 0.25 μg/ml amphotericin B (all from Thermo Fisher Scientific, Waltham, MA). For transfections, cells were seeded into 12-well plates [real-time quantitative PCR (RT-qPCR) and Western blotting] or 10 cm dishes (cell fractionation) and left overnight to recover. The following day, 800 ng (for 12-well plates) of plasmid or empty vector (EV) were transfected into cells using Lipofectamine 2000 (Thermo Fisher Scientific) according to manufacturer’s recommendations. Effects on transcription and translation were assessed after 24 h unless stated otherwise. The plasmids used were: AIP-Myc-pcDNA, AIP_R304*-Myc-pcDNA (27), pcDNA-PRKAR1A (38), pCMV-PRKACA (OriGene, Rockville, MD), PRKAR1A-HA, PRKACA-HA, HSC70-HA and HSP90-HA [all cloned into the pSF-CMV-NH2-HA-EKT-NcoI vector (Oxford Genetics OG93, Oxford, UK)], pVenus-PRKAR1A, pCerulean-PRKACA (38), a firefly luciferase reporter vector driven by a cAMP responsive element (CRE, pGL4.29[luc2P/CRE/Hygro], Promega, Madison, WI) and a Renilla luciferase reporter control vector (pRL-SV40, Promega). For siRNA experiments, 2 μl of 20 μM Aip siRNA or non-targeting control siRNA (Thermo Fisher Scientific) were transfected analogous to overexpression experiments (above, same paragraph). For stimulations, cells were seeded into 12-well plates and left overnight to recover. Cells were then stimulated with PKI (PKA inhibitor 6–22 amide, Merck Millipore, Darmstadt, Germany) or forskolin (fsk, Sigma-Aldrich, St. Louis, MO) for 8 h. Cells were then lysed and subjected to RNA isolation. For protein stability experiments, cells were seeded into 12-well plates and left overnight to recover. Cells were then transfected and left to recover for 6 h, before placing them in serum-free medium. After 4 h, the translation inhibitor cycloheximide (CHX, 5 μg/ml, Sigma-Aldrich), and the proteasome inhibitor MG132 (1 μM, Sigma-Aldrich) or vehicle (dimethyl sulfoxide, DMSO, Sigma-Aldrich) were added. After an incubation period of 14 h, cells were lysed and subjected to Western blotting. RNA isolation, reverse transcription and RT-qPCR RNA was isolated using the RNeasy kit (Qiagen) including on-column DNase I treatment. About 500 ng of RNA were reverse transcribed using Superscript III (Life Technologies) and cDNA levels were quantified by multiplex RT-qPCR using TaqMan gene expression assays for rat Aip (Rn00597273_m1) along with beta-Actin (Actb, Rn00667869_m1) on a ViiA Real-Time PCR system (all from Life Technologies) according to manufacturer’s instructions. Relative gene expression was calculated using the ΔΔCt method. Confocal immunofluorescence GH3 cells were seeded onto poly-l-lysine (Sigma-Aldrich) coated glass chamber slides (Thermo Fisher Scientific) and left overnight to recover. They were then placed in low-serum medium (0.5% FBS) for 24 h before stimulations. Cells were pre-incubated with the competitive PKA inhibitor H-89 (Sigma-Aldrich) at 10 μM or vehicle for 1 h before stimulation with the adenylyl cyclase activator forskolin (fsk, Sigma-Aldrich) at 10 μM or vehicle for 30 min. Cells were then washed twice with phosphate-buffered saline (PBS, Thermo Fisher Scientific) and fixed in 4% paraformaldehyde (Sigma-Aldrich) for 15 min. They were blocked in 10% horse serum (Jackson ImmunoResearch Labs, West Grove, PA) in phosphate-buffered saline (PBS) containing 0.1% Triton X-100 (Sigma-Aldrich) for 1 h and then incubated with primary antibodies anti-AIP (1:200, Novus Biologicals, Littleton, CO) and anti-PRKAR1A (1:100, Abcam, Cambridge, UK) or anti-PRKACA (1:50, Santa Cruz Biotechnology, Dallas, TX) in 10% horse serum for 2 h at room temperature. Slides were washed and incubated with secondary antibodies Alexa Fluor 488 donkey anti-mouse and Alexa Fluor 555 donkey anti-rabbit (1:2000, Thermo Fisher Scientific) in PBS for 1 h at room temperature. Slides were washed and mounted using Prolong Gold mount with DAPI (4′, 6-diamidino-2-phenylindole, Thermo Fisher Scientific). Negative controls without primary antibody were included in every experiment. Confocal immunofluorescence experiments were performed as described previously (38). Briefly, images were captured using a Zeiss LSM 510 inverted confocal microscope with a 63× Zeiss plan-apochromat oil immersion objective (NA 1.4) or a 40× Zeiss plan-neofluar oil immersion objective (NA 1.3), as stated, and a 488-nm Argon ion laser with a 505–530 nm band pass emission filter, a 543-nm helium-neon laser and a 560–615 emission filter, and a 405-nm solid-state laser with a 420–470 nm emission filter, with the LSM 510 software. Colocalization was quantified in images taken from 10 separate visual fields using the Coloc2 plugin in ImageJ (National Institutes of Health, Bethesda, MD) to calculate Pearson’s R above threshold. Differences between means were assessed by independent samples Student’s t-tests. Fluorescence resonance energy transfer by acceptor photobleaching GH3 cells were seeded onto poly-l-lysine (Sigma-Aldrich) coated glass chamber slides (Thermo Fisher Scientific) and left overnight to recover. pVenus-R1α (acceptor), pCerulean-Cα (donor) and either AIP or EV were transfected and left for 24 h. About 1 h prior to FRET, cells were stimulated with either fsk or vehicle. Experiments were performed on the same microscope and with the same software used for confocal immunofluorescence (see above). Cα-Cerulean was imaged with the 405-nm laser, R1α-Venus with the 514-nm laser. Using a custom region of interest, R1a-Venus in one cell was bleached with the 514-nm laser at 100% transmission until the overall intensity dropped to between 80% and 50% of pre-bleach values. The intensity of Cα-Cerulean and R1α-Venus were concurrently measured in a neighboring cell not exposed to the bleaching laser to ensure that no significant bleaching of either proteins took place during imaging. The R1α-Cα interaction was calculated by the difference in Cerulean intensity pre- versus post-bleaching. In each experiment, 15–20 bleaches on different cells were performed. Results shown are the combined results of three separate experiments. Cell fractionation GH3 cells were seeded onto 10-cm cell culture dishes and the following day, AIP-Myc or EV (pcDNA) were transfected along with Cα-HA and R1α-HA. After 24 h, cells were counted and equal numbers were lysed and cell fractionation was performed using the subcellular protein fractionation kit (Thermo Fisher Scientific) according to manufacturer’s instructions. Lysates were subjected to Western blotting (described below). Western blotting Total lysates were prepared from cell pellets using the M-PER protein extraction reagent (Thermo Fisher Scientific) containing protease inhibitors (Complete, Roche, Basel, Switzerland). Protein concentrations were determined by BCA protein assay (Thermo Fisher Scientific) and 15 μg of lysates were subjected to SDS–PAGE and Western Blotting by standard laboratory methods. Membranes were probed with anti-AIP (Novus Biologicals) (10,33,39), anti-HA (Abcam), anti-Myc (Sigma-Aldrich), anti-GAPDH (Santa Cruz), anti-Histone H3 (Abcam), anti-TORC (Merck-Millipore, Billerica, MA), anti-HPRT (Santa Cruz), anti-e-cadherin (Abcam) or anti-α-Tubulin (Abcam), as stated, followed by appropriate HRP-coupled secondary antibodies (Jackson ImmunoResearch). Signals were detected by enhanced chemiluminescence and visualized on a ChemiDoc imaging system (Bio-Rad Laboratories, Hercules, CA). Co-immunoprecipitations For co-IPs, HEK293 cells were seeded onto 10-cm cell culture dishes and the following day, different combinations of AIP-Myc, R1α-HA, Cα-HA, untagged R1α-pcDNA or Cα-pCMV, HSP90-HA or HSC70-HA, as stated, were co-transfected. After 24 h, crosslinking was performed using 16% paraformaldehyde for 15’. Co-IPs were performed using the Pierce Co-Immunoprecipitation Kit (Thermo Fisher Scientific), in which anti-Myc or anti-HA antibodies (as stated) were covalently linked to an amine-reactive resin, according to manufacturer’s instructions. Briefly, cells were lysed, the lysate was pre-cleared using negative control beads, and some material was retained for the input control lanes, after which co-IPs with the antibody-coupled resin (10 μg antibody and 10 μl resin per reaction) were performed at 4°C overnight. Resin coupled to IgG from the same species as the IP antibody (mouse or rabbit IgG, both from Sigma-Aldrich) was used as a negative control. The following day, beads were washed three times and the bound fraction was eluted. The elution fractions were subjected to Western blotting alongside the pre-IP input lysate. PKA activity PKA activity was determined by the measurement of 32P incorporation as described previously (40). Briefly, cells were lysed and homogenized in 10 mM Tris–HCl, pH 7.0, 1 mM ethylenediaminetetraacetic acid, 0.1 mM dithiothreitol and protease inhibitor cocktail I (Merck, Kenilworth, NJ) with or without 5 µM cAMP. Samples were incubated at 37°C for 15 min after which the reaction was stopped on ice. Samples were spotted onto phosphocellulose filters, washed in phosphoric acid and dried. Analysis was performed by liquid scintillation counting. CRE-activity, double luciferase assay GH3 cells were seeded into 12-well plates and left overnight to recover. The following day, cells were transfected with a firefly luciferase reporter vector driven by CRE (CRE-Luc) and a Renilla luciferase control reporter vector (R-Luc) as well as siRNA, as stated. Firefly luciferase and Renilla luciferase activities were assessed with a dual luciferase reporter assay system (Promega) according to manufacturer’s instructions on a FLUOstar Omega luminometer (BMG Labtech, Ortenberg, Germany). PDE inhibitors used were: the nonspecific PDE inhibitor 3-isobutyl-1-methylxanthine (IBMX, 10 μM, Sigma-Aldrich), the PDE2 inhibitor erythro-9-amino-β-hexyl-α-methyl-9H-ethanol hydrochloride (EHNA, 10 μM, Sigma-Aldrich), the PDE4 inhibitor Rolipram (10 μM, Sigma-Aldrich), the PDE2A inhibitor BAY 60–7550 (50 nM, Cayman Chemical, Ann Arbor, MI) and the PDE11 inhibitor BC 11–38 (20 µM, R&D Systems, Minneapolis, MN) (41). Statistics Bar charts show means ±S.E.M. Differences between means were computed by independent samples Student’s t-tests, one-way ANOVA or two-way ANOVA followed by post hoc testing, as appropriate, using the Prism software (GraphPad, La Jolla, CA). Supplementary Material Supplementary Material is available at HMG online. Acknowledgements The project was supported by the Intramural Research Program, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health (grant number Z01-HD008920), and by the Austrian Science Fund (FWF) (Erwin Schrödinger Fellowship, grant number J3482-B13) to M.H.S.‐R. We would like to thank Prof. Márta Korbonits (Barts and the London School of Medicine and Dentistry, Queen Mary University of London) for providing the AIP, HSC70 and HSP90 plasmids, as well as Dr Forbes D. Porter and Dr Anthony Cougnoux (NICHD, NIH) for technical advice on co-immunoprecipitations. Finally, we thank Dr Vincent Schram (NICHD, NIH) and the NICHD Microscopy and Imaging Core for technical advice in co-immunofluorescence and FRET experiments and Prof. Charles Hoffman (Biology Department, Boston College) for providing the PDE11 inhibitor BC11–38. Conflict of Interest statement. None declared. References 1 Vierimaa O. , Georgitsi M. , Lehtonen R. , Vahteristo P. , Kokko A. , Raitila A. , Tuppurainen K. , Ebeling T.M. , Salmela P.I. , Paschke R. ( 2006 ) Pituitary adenoma predisposition caused by germline mutations in the AIP gene . Science , 312 , 1228 – 1230 . Google Scholar CrossRef Search ADS PubMed 2 Daly A.F. , Vanbellinghen J.F. , Khoo S.K. , Jaffrain-Rea M.L. , Naves L.A. , Guitelman M.A. , Murat A. , Emy P. , Gimenez-Roqueplo A.P. , Tamburrano G. et al. ( 2007 ) Aryl hydrocarbon receptor-interacting protein gene mutations in familial isolated pituitary adenomas: analysis in 73 families . J. Clin. Endocrinol. Metab ., 92 , 1891 – 1896 . Google Scholar CrossRef Search ADS PubMed 3 Igreja S. , Chahal H.S. , King P. , Bolger G.B. , Srirangalingam U. , Guasti L. , Chapple J.P. , Trivellin G. , Gueorguiev M. , Guegan K. et al. ( 2010 ) Characterization of aryl hydrocarbon receptor interacting protein (AIP) mutations in familial isolated pituitary adenoma families . Hum. Mutat ., 31 , 950 – 960 . Google Scholar CrossRef Search ADS PubMed 4 Hernandez-Ramirez L.C. , Gabrovska P. , Denes J. , Stals K. , Trivellin G. , Tilley D. , Ferrau F. , Evanson J. , Ellard S. , Grossman A.B. et al. ( 2015 ) Landscape of familial isolated and young-onset pituitary adenomas: prospective diagnosis in AIP mutation carriers . J. Clin. Endocrinol. Metab ., 100 , E1242 – E1254 . Google Scholar CrossRef Search ADS PubMed 5 Trivellin G. , Korbonits M. ( 2011 ) AIP and its interacting partners . J. Endocrinol ., 210 , 137 – 155 . Google Scholar CrossRef Search ADS PubMed 6 Bolger G.B. , Peden A.H. , Steele M.R. , MacKenzie C. , McEwan D.G. , Wallace D.A. , Huston E. , Baillie G.S. , Houslay M.D. ( 2003 ) Attenuation of the activity of the cAMP-specific phosphodiesterase PDE4A5 by interaction with the immunophilin XAP2 . J. Biol. Chem ., 278 , 33351 – 33363 . Google Scholar CrossRef Search ADS PubMed 7 de Oliveira S.K. , Hoffmeister M. , Gambaryan S. , Muller-Esterl W. , Guimaraes J.A. , Smolenski A.P. ( 2007 ) Phosphodiesterase 2A forms a complex with the co-chaperone XAP2 and regulates nuclear translocation of the aryl hydrocarbon receptor . J. Biol. Chem ., 282 , 13656 – 13663 . Google Scholar CrossRef Search ADS PubMed 8 Nakata A. , Urano D. , Fujii-Kuriyama Y. , Mizuno N. , Tago K. , Itoh H. ( 2009 ) G-protein signalling negatively regulates the stability of aryl hydrocarbon receptor . EMBO Rep ., 10 , 622 – 628 . Google Scholar CrossRef Search ADS PubMed 9 Hernandez-Ramirez L.C. , Trivellin G. , Stratakis C.A. ( 2017 ) Role of phosphodiesterases on the function of aryl hydrocarbon receptor-interacting protein (AIP) in the pituitary gland and on the evaluation of AIP gene variants . Horm. Metab. Res ., 49 , 286 – 295 . Google Scholar CrossRef Search ADS PubMed 10 Formosa R. , Xuereb-Anastasi A. , Vassallo J. ( 2013 ) Aip regulates cAMP signalling and GH secretion in GH3 cells . Endocr. Relat. Cancer , 20 , 495 – 505 . Google Scholar CrossRef Search ADS PubMed 11 Lania A.G. , Mantovani G. , Ferrero S. , Pellegrini C. , Bondioni S. , Peverelli E. , Braidotti P. , Locatelli M. , Zavanone M.L. , Ferrante E. et al. ( 2004 ) Proliferation of transformed somatotroph cells related to low or absent expression of protein kinase a regulatory subunit 1A protein . Cancer Res ., 64 , 9193 – 9198 . Google Scholar CrossRef Search ADS PubMed 12 Vitali E. , Peverelli E. , Giardino E. , Locatelli M. , Lasio G.B. , Beck-Peccoz P. , Spada A. , Lania A.G. , Mantovani G. ( 2014 ) Cyclic adenosine 3′-5′-monophosphate (cAMP) exerts proliferative and anti-proliferative effects in pituitary cells of different types by activating both cAMP-dependent protein kinase A (PKA) and exchange proteins directly activated by cAMP (Epac) . Mol. Cell. Endocrinol ., 383 , 193 – 202 . Google Scholar CrossRef Search ADS PubMed 13 Landis C.A. , Masters S.B. , Spada A. , Pace A.M. , Bourne H.R. , Vallar L. ( 1989 ) GTPase inhibiting mutations activate the alpha chain of Gs and stimulate adenylyl cyclase in human pituitary tumours . Nature , 340 , 692 – 696 . Google Scholar CrossRef Search ADS PubMed 14 Vallar L. , Spada A. , Giannattasio G. ( 1987 ) Altered Gs and adenylate cyclase activity in human GH-secreting pituitary adenomas . Nature , 330 , 566 – 568 . Google Scholar CrossRef Search ADS PubMed 15 Freda P.U. , Chung W.K. , Matsuoka N. , Walsh J.E. , Kanibir M.N. , Kleinman G. , Wang Y. , Bruce J.N. , Post K.D. ( 2007 ) Analysis of GNAS mutations in 60 growth hormone secreting pituitary tumors: correlation with clinical and pathological characteristics and surgical outcome based on highly sensitive GH and IGF-I criteria for remission . Pituitary , 10 , 275 – 282 . Google Scholar CrossRef Search ADS PubMed 16 Valimaki N. , Demir H. , Pitkanen E. , Kaasinen E. , Karppinen A. , Kivipelto L. , Schalin-Jantti C. , Aaltonen L.A. , Karhu A. ( 2015 ) Whole-genome sequencing of growth hormone (GH)-secreting pituitary adenomas . J. Clin. Endocrinol. Metab ., 100 , 3918 – 3927 . Google Scholar CrossRef Search ADS PubMed 17 Akintoye S.O. , Chebli C. , Booher S. , Feuillan P. , Kushner H. , Leroith D. , Cherman N. , Bianco P. , Wientroub S. , Robey P.G. et al. ( 2002 ) Characterization of gsp-mediated growth hormone excess in the context of McCune-Albright syndrome . J. Clin. Endocrinol. Metab ., 87 , 5104 – 5112 . Google Scholar CrossRef Search ADS PubMed 18 Vortmeyer A.O. , Glasker S. , Mehta G.U. , Abu-Asab M.S. , Smith J.H. , Zhuang Z. , Collins M.T. , Oldfield E.H. ( 2012 ) Somatic GNAS mutation causes widespread and diffuse pituitary disease in acromegalic patients with McCune-Albright syndrome . J. Clin. Endocrinol. Metab ., 97 , 2404 – 2413 . Google Scholar CrossRef Search ADS PubMed 19 Kirschner L.S. , Carney J.A. , Pack S.D. , Taymans S.E. , Giatzakis C. , Cho Y.S. , Cho-Chung Y.S. , Stratakis C.A. ( 2000 ) Mutations of the gene encoding the protein kinase A type I-alpha regulatory subunit in patients with the Carney complex . Nat. Genet ., 26 , 89 – 92 . Google Scholar CrossRef Search ADS PubMed 20 Robinson-White A. , Hundley T.R. , Shiferaw M. , Bertherat J. , Sandrini F. , Stratakis C.A. ( 2003 ) Protein kinase-A activity in PRKAR1A-mutant cells, and regulation of mitogen-activated protein kinases ERK1/2 . Hum. Mol. Genet ., 12 , 1475 – 1484 . Google Scholar CrossRef Search ADS PubMed 21 Skroblin P. , Grossmann S. , Schafer G. , Rosenthal W. , Klussmann E. ( 2010 ) Mechanisms of protein kinase A anchoring . Int. Rev. Cell Mol. Biol ., 283 , 235 – 330 . Google Scholar CrossRef Search ADS PubMed 22 Stojilkovic S.S. , Kretschmannova K. , Tomic M. , Stratakis C.A. ( 2012 ) Dependence of the excitability of pituitary cells on cyclic nucleotides . J. Neuroendocrinol ., 24 , 1183 – 1200 . Google Scholar CrossRef Search ADS PubMed 23 Peverelli E. , Ermetici F. , Filopanti M. , Elli F.M. , Ronchi C.L. , Mantovani G. , Ferrero S. , Bosari S. , Beck-Peccoz P. , Lania A. et al. ( 2009 ) Analysis of genetic variants of phosphodiesterase 11A in acromegalic patients . Eur. J. Endocrinol ., 161 , 687 – 694 . Google Scholar CrossRef Search ADS PubMed 24 Bolger G.B. , Bizzi M.F. , Pinheiro S.V. , Trivellin G. , Smoot L. , Accavitti M.A. , Korbonits M. , Ribeiro-Oliveira A. Jr ( 2016 ) cAMP-specific PDE4 phosphodiesterases and AIP in the pathogenesis of pituitary tumors . Endocr. Relat. Cancer , 23 , 419 – 431 . Google Scholar CrossRef Search ADS PubMed 25 Persani L. , Borgato S. , Lania A. , Filopanti M. , Mantovani G. , Conti M. , Spada A. ( 2001 ) Relevant cAMP-specific phosphodiesterase isoforms in human pituitary: effect of Gs(alpha) mutations . J. Clin. Endocrinol. Metab ., 86 , 3795 – 3800 . Google Scholar PubMed 26 Hernández-Ramírez L.C. , Morgan R.M.L. , Barry S. , D’Acquisto F. , Prodromou C. , Korbonits M. ( 2018 ) Multi-chaperone function modulation and association with cytoskeletal proteins are key features of the function of AIP in the pituitary gland . Oncotarget , 9 , 9177 – 9198 . Google Scholar CrossRef Search ADS PubMed 27 Leontiou C.A. , Gueorguiev M. , van der Spuy J. , Quinton R. , Lolli F. , Hassan S. , Chahal H.S. , Igreja S.C. , Jordan S. , Rowe J. et al. ( 2008 ) The role of the aryl hydrocarbon receptor-interacting protein gene in familial and sporadic pituitary adenomas . J. Clin. Endocrinol. Metab ., 93 , 2390 – 2401 . Google Scholar CrossRef Search ADS PubMed 28 Morgan R.M. , Hernandez-Ramirez L.C. , Trivellin G. , Zhou L. , Roe S.M. , Korbonits M. , Prodromou C. ( 2012 ) Structure of the TPR domain of AIP: lack of client protein interaction with the C-terminal alpha-7 helix of the TPR domain of AIP is sufficient for pituitary adenoma predisposition . PLoS One , 7 , e53339. Google Scholar CrossRef Search ADS PubMed 29 Hernandez-Ramirez L.C. , Martucci F. , Morgan R.M. , Trivellin G. , Tilley D. , Ramos-Guajardo N. , Iacovazzo D. , D’Acquisto F. , Prodromou C. , Korbonits M. ( 2016 ) Rapid proteasomal degradation of mutant proteins is the primary mechanism leading to tumorigenesis in patients with missense AIP mutations . J. Clin. Endocrinol. Metab ., 101 , 3144 – 3154 . Google Scholar CrossRef Search ADS PubMed 30 Tuominen I. , Heliovaara E. , Raitila A. , Rautiainen M.R. , Mehine M. , Katainen R. , Donner I. , Aittomaki V. , Lehtonen H.J. , Ahlsten M. et al. ( 2015 ) AIP inactivation leads to pituitary tumorigenesis through defective Galphai-cAMP signaling . Oncogene , 34 , 1174 – 1184 . Google Scholar CrossRef Search ADS PubMed 31 Oesch-Bartlomowicz B. , Oesch F. ( 2009 ) Role of cAMP in mediating AHR signaling . Biochem. Pharmacol ., 77 , 627 – 641 . Google Scholar CrossRef Search ADS PubMed 32 Oesch-Bartlomowicz B. , Huelster A. , Wiss O. , Antoniou-Lipfert P. , Dietrich C. , Arand M. , Weiss C. , Bockamp E. , Oesch F. ( 2005 ) Aryl hydrocarbon receptor activation by cAMP vs. dioxin: divergent signaling pathways . Proc. Natl. Acad. Sci. U. S. A ., 102 , 9218 – 9223 . Google Scholar CrossRef Search ADS PubMed 33 Heliovaara E. , Raitila A. , Launonen V. , Paetau A. , Arola J. , Lehtonen H. , Sane T. , Weil R.J. , Vierimaa O. , Salmela P. et al. ( 2009 ) The expression of AIP-related molecules in elucidation of cellular pathways in pituitary adenomas . Am. J. Pathol ., 175 , 2501 – 2507 . Google Scholar CrossRef Search ADS PubMed 34 Raitila A. , Lehtonen H.J. , Arola J. , Heliovaara E. , Ahlsten M. , Georgitsi M. , Jalanko A. , Paetau A. , Aaltonen L.A. , Karhu A. ( 2010 ) Mice with inactivation of aryl hydrocarbon receptor-interacting protein (Aip) display complete penetrance of pituitary adenomas with aberrant ARNT expression . Am. J. Pathol ., 177 , 1969 – 1976 . Google Scholar CrossRef Search ADS PubMed 35 Jaffrain-Rea M.L. , Angelini M. , Gargano D. , Tichomirowa M.A. , Daly A.F. , Vanbellinghen J.F. , D’Innocenzo E. , Barlier A. , Giangaspero F. , Esposito V. et al. ( 2009 ) Expression of aryl hydrocarbon receptor (AHR) and AHR-interacting protein in pituitary adenomas: pathological and clinical implications . Endocr. Relat. Cancer , 16 , 1029 – 1043 . Google Scholar CrossRef Search ADS PubMed 36 Lecoq A.L. , Viengchareun S. , Hage M. , Bouligand J. , Young J. , Boutron A. , Zizzari P. , Lombes M. , Chanson P. , Kamenicky P. ( 2016 ) AIP mutations impair AhR signaling in pituitary adenoma patients fibroblasts and in GH3 cells . Endocr. Relat. Cancer , 23 , 433 – 443 . Google Scholar CrossRef Search ADS PubMed 37 Kirschner L.S. , Yin Z. , Jones G.N. , Mahoney E. ( 2009 ) Mouse models of altered protein kinase A signaling . Endocr. Relat. Cancer , 16 , 773 – 793 . Google Scholar CrossRef Search ADS PubMed 38 Mavrakis M. , Lippincott-Schwartz J. , Stratakis C.A. , Bossis I. ( 2006 ) Depletion of type IA regulatory subunit (RIalpha) of protein kinase A (PKA) in mammalian cells and tissues activates mTOR and causes autophagic deficiency . Hum. Mol. Genet ., 15 , 2962 – 2971 . Google Scholar CrossRef Search ADS PubMed 39 Chahal H.S. , Trivellin G. , Leontiou C.A. , Alband N. , Fowkes R.C. , Tahir A. , Igreja S.C. , Chapple J.P. , Jordan S. , Lupp A. et al. ( 2012 ) Somatostatin analogs modulate AIP in somatotroph adenomas: the role of the ZAC1 pathway . J. Clin. Endocrinol. Metab ., 97 , E1411 – E1420 . Google Scholar CrossRef Search ADS PubMed 40 Nesterova M.V. , Sashchenko L.P. , Vasiliev V.Y. , Severin E.S. ( 1975 ) A cyclic adenosine 3′, 5′-monophosphate-dependent histone kinase from pig brain. Purification and some properties of the enzyme . Biochim. Biophys. Acta , 377 , 271 – 281 . Google Scholar CrossRef Search ADS PubMed 41 Ceyhan O. , Birsoy K. , Hoffman C.S. ( 2012 ) Identification of biologically active PDE11-selective inhibitors using a yeast-based high-throughput screen . Chem. Biol ., 19 , 155 – 163 . Google Scholar CrossRef Search ADS PubMed Published by Oxford University Press 2018. This work is written by US Government employees and is in the public domain in the US. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Human Molecular GeneticsOxford University Press

Published: May 2, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off