Endogenous Protein Kinase Inhibitor (cid:1) Terminates Immediate-early Gene Expression Induced by cAMP-dependent Protein Kinase (PKA) Signaling TERMINATION DEPENDS ON PKA INACTIVATION RATHER THAN PKA EXPORT FROM THE NUCLEUS

,

Many genes are stimulated by cAMP-dependent protein kinase (PKA) 1 signaling in an immediate-early fashion characterized by rapid but transient induction of gene transcription (1,2). This transient gene transcription is associated with transient elevation of cAMP levels, transient PKA activation, and transient phosphorylation of both cAMP-responsive element-binding protein (CREB) and the related transcription factor, activating transcription factor-1 (ATF-1) (1, 3). A well studied example is the stimulation of immediate-early genes such as interleukin-6 (IL-6) and c-fos in osteoblasts by parathyroid hormone (PTH) (2-7). These immediate-early genes are induced in response to PTH by PKA-dependent phosphorylation of transcription factors, such as CREB and CCAAT/ enhancer binding protein (3,4,7,8). Because these immediateearly genes mediate both the catabolic and anabolic effects of PTH (9), elucidation of the regulation of expression of these genes will substantially increase our understanding of the mechanism responsible for fine-tuning the balance between the catabolic and anabolic effects.
Transient regulation of gene transcription is most likely due to termination of one or more step(s) in the relevant signal transduction pathway. Consistent with this, multiple mechanisms contribute to precise regulation of both PKA signaling and the changes in gene expression induced by this signaling pathway (1,3,10). We have previously studied regulation of immediate-early gene expression following activation of PKA signaling by PTH or the ␤-adrenergic agonist, isoproterenol (ISO), in osteoblasts and fibroblasts (3). Those studies demonstrate that the primary mechanism responsible for termination of immediate-early gene expression acts downstream of receptor desensitization, adenylyl cyclase activation, and cAMP degradation (3). Possible mechanisms therefore include inactivation of PKA itself (11), dephosphorylation or proteolysis of downstream transcription factors such as CREB (12)(13)(14), activation of inhibitory transcription factors such as inducible cAMP early repressor (15), histone deacetylation (13,16), and co-activator methylation (17). We have focused on inactivation of PKA because it is the only one of these mechanisms that could account not only for termination of immediate-early gene transcription but also for the termination of both PKA activity and transcription factor phosphorylation that occur following activation of PKA signaling (3).
In the cytoplasm, activity of the catalytic subunit of PKA is terminated in response to reduced cAMP levels by its reassociation with regulatory subunits to form an inactive holoenzyme composed of two regulatory and two catalytic subunits (11). In contrast, the best candidates for terminating nuclear PKA catalytic subunit activity are protein kinase inhibitors (PKIs) (3,11). The PKIs are a family of proteins that act as pseudosubstrates for PKA by binding to the PKA catalytic subunits and inhibiting their enzymatic activity (11,18). Three PKI genes (PKI␣, PKI␤, and PKI␥) have been identified (19). When PKI is overexpressed, it accumulates in the nucleus (20), binds and inactivates the catalytic subunits of PKA (11,(21)(22)(23), and transports them back to the cytoplasm (21,(23)(24)(25)(26)(27). This transport of PKA catalytic subunits depends on a leucine-rich nuclear export signal in PKI (26). Despite the large number of studies examining purified or overexpressed PKI (11, 18, 20 -29), little is known about the physiological role of endogenous PKI. In fact, the PKI␣ and PKI␤ genes have been ablated by homologous recombination with little detectable effect on phenotype, even in PKI␣/PKI␤ double knock-out mice (30,31). This result appears to be because of compensation by increased expression of other PKI family members and/or the regulatory subunits of PKA (30). Antisense oligonucleotide studies have implicated PKI␣ in regulation of synaptic activity in the rat hippocampus (32) and in development of left-right asymmetry in chick embryos (33).
In this study, we used both siRNA and antisense knockdown strategies to examine the role of PKI␥ in PKA signaling activated by PTH or ISO. We found that endogenous PKI␥ was required for efficient termination of immediate-early gene expression following activation of PKA signaling. Endogenous PKI␥ was also required for export of PKA catalytic subunits from the nucleus back to the cytoplasm. However, inactivation of PKA in the nucleus by endogenous PKI␥ was sufficient to terminate signaling, and nuclear export was not required. These results were the first in any cell type showing that endogenous levels of PKI regulate PKA signaling.

EXPERIMENTAL PROCEDURES
Cell Culture-All cell culture media, supplements, and stimulators were screened for endotoxin contamination using the high sensitivity version of the colorimetric Limulus amoebocyte lysate assay (QCL-1000; Whittaker Bioproducts, Walkersville, MD) as described previously (34). Endotoxin levels were Ͻ0.004 endotoxin units/ml for the concentration of each stimulator used in the experiments. Rat ROS 17/2.8 osteoblastic osteosarcoma cells (35), murine MC3T3-E1 osteoblastic cells (36), and murine NIH3T3 fibroblastic cells (ATCC CRL 1658) were maintained as described previously (3). All cell lines were cultured at 37°C in a humidified atmosphere containing 95% air and 5% CO 2 and routinely passaged every 3 or 4 days.
siRNA Experiments-Rat PKI␥ cDNA from ROS 17/2.8 cells was amplified by PCR using primers (5Ј-GAAGAGATGCAGGCAGGA-3Ј and 5Ј-AGAGGTACAGCAGGTCTC-3Ј) based on the murine sequence (GenBank TM accession code U97170). PCR products were sequenced and the rat PKI␥ sequence was deposited in GenBank TM (accession code AY150308). Five siRNA target sequences directed against the rat PKI␥ coding region were selected following the recommendations of Tuschl and co-workers (37). All of these targets therefore are 21 nucleotides in length, are preceded by 5Ј-AA, avoid the first 100 nucleotides of the coding region, contain ϳ50% G/C content, and contain at least three nucleotide mismatches with all other sequences in the data base. siRNA duplexes directed at each PKI␥ target were prepared by in vitro transcription. siRNA duplexes targeting firefly luciferase were also prepared for use as controls as recommended (37)(38)(39). In vitro transcription was performed using the Silencer TM siRNA construction kit (Ambion, TX) and sense and antisense oligonucleotides obtained from Integrated DNA Technologies, Inc. (Coralville, IA).
For transfection of siRNA duplexes, ROS 17/2.8 cells were seeded into 6-well plates at a density of 5 ϫ 10 5 cells/cm 2 in F-12 nutrient mixture (Ham's F-12 medium) (Invitrogen) with 10% fetal bovine serum (Hyclone, Logan, UT) without antibiotics. Four different transfection reagents, including siPORT amine and siPORT lipid (Ambion, Austin, TX), FuGENE 6 (Roche Applied Science), and Lipofectamine 2000 (Invitrogen), were tested during screening experiments. The most profound knockdown of glyceraldehyde-3-phosphate dehydrogenase mRNA was achieved with Lipofectamine 2000 (data not shown). This reagent was therefore used for all further siRNA transfections. Antisense Experiments-cDNA from ROS 17/2.8 cells encoding nucleotides 3-271 of rat PKI␥ (GenBank TM accession code AY150308) was amplified by PCR, cloned into pCR2.1-TOPO vector (Invitrogen), and subcloned into pTRE expression vectors (Clontech, Palo Alto, CA). Plasmid clones containing either sense PKI␥ mRNA (pTRE-PKI␥-S) or antisense PKI␥ mRNA (pTRE-PKI␥-AS) were selected by restriction mapping and sequencing (Case Western Reserve University DNA sequencing core facility). Previously established stable ROS pTet-Off cells (3) were then transfected with pTRE-PKI␥-S, pTRE-PKI␥-AS, or control pTRE-Luc (Clontech) and selected with 20 g/ml of hygromycin B (Sigma) to obtain double stable transfectant clones. In this system (40), withdrawal of tetracycline (Sigma) from the culture medium induces the expression of sense or antisense PKI␥ mRNAs. Double stable transfectant clones were seeded into 60-mm dishes at a density of 2.5 ϫ 10 4 cells/cm 2 and incubated for 3 days with tetracycline until they reached confluence. The cells were then cultured with or without tetracycline after being washed three times with PBS. After 48 h, the cells were incubated with 100 nM bPTH (1-34) in serum-free Ham's F-12 medium containing 0.1% BSA for the indicated periods of time.
Leptomycin B Experiments-ROS 17/2.8 cells were seeded into 60-mm dishes at a density of 2.5 ϫ 10 4 cells/cm 2 . 24 h after reaching confluence, the cells were preincubated with or without 10 ng/ml leptomycin B (41) obtained from Minoru Yoshida (RIKEN, Wako, Japan) in serum-free Ham's F-12 medium containing 0.1% BSA. After 1 h, 100 nM bPTH  or vehicle control (1 M acetic acid) were added and incubation continued for the indicated periods of time.
Western Analysis-For total cell lysate preparation, cells were washed twice with ice-cold PBS and lysed with SDS lysis buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, and 50 mM dithiothreitol) without bromphenol blue. Nuclear and cytosolic fractions were prepared as described (42). Cells were washed twice with ice-cold PBS and scraped into lysis buffer (10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl 2 , 0.3% Nonidet P-40) containing protease inhibitor mixture (Roche Applied Science) and phosphatase inhibitors (1 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, and 1 mM sodium orthovanadate). Cells were incubated on ice for 3 min followed by centrifugation at 500 ϫ g for 5 min at 4°C. The supernatants contained the cytosolic fraction. Pellets containing cell nuclei were washed twice with lysis buffer without Nonidet P-40 and lysed with SDS lysis buffer. Nuclear and total cell lysates were sonicated on ice for 5 s three times to shear DNA and centrifuged at 12,000 ϫ g for 10 min at 4°C. The protein concentrations of all lysates were measured using the Bradford dye binding assay (Bio-Rad Laboratories, Hercules, CA). Equal amounts of protein were electrophoresed on SDS-PAGE gels (BioWhittaker Molecular Applications, Rockland, ME). Proteins were electrotransferred onto polyvinylidene difluoride membranes (Bio-Rad) at 24 V for 1.5 h in 25 mM Tris-HCl, pH 8.5, 0.2 M glycine, 20% methanol. The membranes were probed with affinity-purified rabbit polyclonal antibodies specific for AKT, CREB and phosphoCREB/phosphoATF-1 (Cell Signaling Technology, Beverly, MA), PKA␣, PKA␤, and PKA␥ (Santa Cruz Biotechnology, Santa Cruz, CA), or PKI␥ (obtained from Michael Uhler, University of Michigan (21)). The PKI␥ antibody was affinity-purified using a PKI␥-maltose-binding protein fusion protein (21) cross-linked to UltraLink beads via azlactone coupling chemistry (UltraLink TM immobilization kit; Pierce). Detection was performed using horseradish peroxidase-conjugated goat anti-rabbit IgG (Cell Signaling Technology) and enhanced chemiluminescence (Amersham Biosciences).
PKA Activity Assays-Nuclear extracts were prepared for PKA as-says as described previously (3). PKA activity was measured using Kemptide as a substrate. Briefly, 10 l of each nuclear extract was incubated in duplicate for 5 min at 30°C with [␥- 32  Reactions were terminated by spotting 20 l onto phosphocellulose discs, and peptide-incorporated 32 P was determined by scintillation counting. PKA activity was calculated by subtracting activity measured in the presence of PKI from total activity of the nuclear extract. PKA activity was normalized to protein concentration as determined by Bradford dye binding assay. Statistical Analysis-All presented RT-PCR and Western blot results are representative of at least three independent experiments. All quantitative data are presented as the mean Ϯ S.E. of all available experiments (n ϭ 3). Symbols without error bars represent S.E. smaller than the symbol. Statistical analyses were by analysis of variance with Fisher's protected least significant difference post hoc tests.

PKI␥ mRNA Is Strongly Expressed in ROS 17/2.8, MC3T3-E1, and NIH3T3 Cells, whereas Little or No mRNA Encoding the Other Family Members (PKI␣ and PKI␤) Is Expressed-
Each PKI family member has a unique pattern of tissue expression (19,21,29). We examined expression of mRNAs encoding each family member in rat ROS 17/2.8 osteoblastic cells (Fig. 1, A and B), murine MC3T3-E1 osteoblastic cells (Fig. 1C), and murine NIH3T3 fibroblastic cells (Fig. 1D). In all cases, PKI␥ mRNA was strongly expressed, whereas little or no mRNA encoding PKI␣ or PKI␤ was detectable (Fig. 1, A-D). The pattern of expression was not altered by activation of PKA signaling (Fig. 1, B-D). PKI␥ is by far the least studied member of the PKI family, with only four citations in the literature (19,21,28,43).
PKI␥ Knockdown by siRNA Substantially Delays Termination of Nuclear PKA Activity, CREB Phosphorylation, and Immediate-early Gene Expression following Stimulation with PTH-Having shown that PKI␥ is the primary member of the PKI family in osteoblastic cells, we performed knockdown experiments to determine whether PKI␥ is required for termination of PKA signaling in these cells. For this purpose, rat PKI␥ cDNA was sequenced (GenBank TM AY150308). We found that the rat PKI␥ coding region is 97.0 and 89.6% identical at the nucleotide level to murine and human PKI␥. At the amino acid level, rat PKI␥ is 100 and 90.8% identical to murine and human PKI␥, respectively. Like PKI␥ in other species (19,21), the rat PKI␥ sequence also contains the pseudosubstrate motif and nuclear export signal.
Five siRNA targets were selected in the coding region of rat PKI␥. Preliminary experiments showed that maximal knockdown was achieved with a mixture of the three siRNA duplexes depicted in Fig. 2A. The extent of knockdown of PKI␥ by siRNA at both the mRNA and protein levels is shown in Fig. 2B, panels 1 and 2, respectively) (see also Figs. 3A and 4A). Importantly, knockdown of PKI␥ in these relatively short experiments did not result in compensatory up-regulation of PKI␣ or PKI␤ mRNAs (Fig. 2B, panels 3 and 4) (see also Figs. 3A, 4A, and 5B). To assess whether endogenous PKI␥ is required for termination of PKA signaling, we compared time course experiments in cultures stimulated with PTH (Fig. 2) or ISO (Fig. 3) following transfection either with PKI␥ siRNA duplexes or with control siRNAs that target firefly luciferase. PKI␥ knockdown by siRNA substantially delayed termination of nuclear PKA activity (Fig. 2C), phosphorylation of the transcription factors CREB and ATF-1 (Figs. 2D and 3B), and expression of the immediate-early genes IL-6 and c-fos (Figs. 2E and 3C).
The experiments just described compared cultures treated with PKI␥ siRNA duplexes to cultures treated with an equal concentration of siRNA duplexes that targeted an irrelevant gene as a control for possible nonspecific effects of siRNA (39,

FIG. 1. PKI␥ mRNA is strongly expressed in ROS 17/2.8 (A and B), MC3T3-E1 (C), and NIH3T3 (D) cells, whereas little or no mRNA encoding the other family members (PKI␣ and PKI␤) is expressed (A-D).
Positive controls are total RNA from rat (A and B) or murine (C and D) tissues that express PKI␣ (skeletal muscle) and PKI␤ (testes). B-D, cultures were incubated for 1 or 4 h without stimulation (C), with 100 nM PTH with or without 100 M 3-isobutyl-1-methylxanthine (IBMX), 10 M ISO with or without 100 M IBMX, or 10 M forskolin (FSK). All cultures also received a mixture of vehicle controls, such that all contained 1 M acetic acid, 0.2% dimethyl sulfoxide, and 0.1 mM NaOH. 44). Two additional approaches were used to confirm that the PKI␥ siRNA results are due to specific effects of PKI␥ knockdown. First, we asked whether PKI␥ knockdown by siRNA affects expression of IL-6 and c-fos mRNAs induced specifically by PKA signaling. For this purpose, we compared the effect of PKI␥ siRNA on gene expression induced by an agent that induces IL-6 and c-fos expression with a time course similar to that induced by PTH but through PKA-independent signaling pathways. TNF␣ was used for this purpose because it stimulates gene expression primarily through NFB and mitogenactivated protein kinase pathways (45), and we have previously shown that a 5-min exposure to TNF␣ stimulates IL-6 gene expression with a time course that is indistinguishable from that induced by PTH. 2 TNF␣ did not detectably affect CREB or ATF-1 phosphorylation in either control or PKI␥ knockdown cells (Fig. 4B), confirming that TNF␣ does not detectably activate PKA signaling. Importantly, PKI␥ siRNA had no effect on expression of IL-6 or c-fos induced by TNF␣ (Fig. 4C). Documentation that PKI␥ was efficiently knocked down in these experiments is shown at both the mRNA and protein levels (Fig. 4A). Moreover, both transcription factor phosphorylation and immediate-early gene expression were substantially increased in replicate cultures treated with PTH (Fig. 4, B and  C). These results showed that PKI␥ siRNA specifically affects PKA signaling and does not have detectable nonspecific effects either on phosphorylation of CREB or ATF-1 or on expression of IL-6 or c-fos.
An additional approach to confirm that the siRNA results are due to specific effects on PKI␥ was to knock down PKI␥ by an independent method, antisense transfection using the Tet-Off system as depicted in Fig. 5A. In this system, tetracycline withdrawal induces expression of the antisense RNA, leading to degradation of the endogenous mRNA and protein. In two distinct antisense clones, tetracycline withdrawal induced efficient knockdown of endogenous PKI␥ at both the mRNA and protein levels (Fig. 5, B and C, first panels, respectively). Moreover, PKI␥ knockdown by tetracycline withdrawal in the antisense clones substantially delayed termination of transcription factor phosphorylation (Fig. 5C, panel 2) and immediate-early gene expression (Fig. 5D) in cultures treated with PTH. As expected in control clones expressing an irrelevant gene, tetracycline did not induce PKI␥ knockdown (Fig. 5, B and C, first panels) and did not alter either transcription factor phosphorylation (Fig. 5C, panel 2) or immediate-early gene expression (Fig. 5D). Further confirmation of the role of PKI␥ was obtained by examining a sense PKI␥ clone, which exhibited opposite effects compared with the antisense clones. Thus, tetracycline withdrawal increased PKI␥ levels (Fig. 5B, panel 2, and  5C, panel 1) and, therefore, reduced transcription factor phosphorylation (Fig. 5C, panel 2) and immediate-early gene ex-2 J. C. Dai, X. Chen, and E. M. Greenfield, manuscript in preparation. pression (Fig. 5D) in response to PTH. These sense and antisense transfection results, taken together with the siRNA results, strongly supported the hypothesis that endogenous PKI␥ contributes to termination of PKA signaling.
PKA Catalytic Subunit Export from the Nucleus Depends on Endogenous PKI␥ but Is Not Required for Termination of PKA Signaling-All three members of the PKI family contain functional CRM1-dependent, leucine-rich nuclear export signal motifs (21,23,26,27). In contrast, the catalytic subunits of PKA do not contain nuclear export signal motifs themselves (24,25) and overexpressed PKI can transport the catalytic subunits out of the nucleus (21,(23)(24)(25). We therefore asked whether endogenous levels of PKI␥ are involved in nuclear export of PKA catalytic subunits following stimulation by PTH. Consistent with this possibility, PTH stimulation rapidly induces PKI␥ entry into the nucleus followed by its re-export back to the cytoplasm (Fig.  6, compare panels 1 and 2). In cells with intact PKI␥ expression, all three isoforms of the PKA catalytic subunits enter the nucleus and are re-exported back to the cytoplasm with time courses similar to that exhibited by PKI␥ (Fig. 6, left half of panels [3][4][5]. PKI␥ knockdown by siRNA blocks the re-export of the PKA catalytic subunits to the cytoplasm (Fig. 6, right half of panels [3][4][5]. These results demonstrated that export of PKA catalytic subunits from the nucleus depends on endogenous PKI␥. When PKI was highly overexpressed, PKI nuclear export was not required to inhibit gene expression (23). However, that study also shows that nuclear export may contribute to inhibition of gene expression when PKI is expressed at lower levels. Together, those results highlighted the importance of examin-ing the mechanisms of PKI␥ action in cells expressing endogenous levels of the protein. For this purpose, we inhibited PKI nuclear export using leptomycin B, which binds to CRM1 and specifically inhibits CRM1-dependent nuclear export (41,47). As expected, leptomycin B completely blocked PKI␥ nuclear export (Fig. 7A, panel 1). Leptomycin B also completely blocked nuclear export of the catalytic subunits of PKA (Fig. 7A, panels  2-4). In contrast, leptomycin B had little or no effect on termination of nuclear PKA activity (Fig. 7B), transcription factor phosphorylation (Fig. 7C), or gene expression (Fig. 7D). Thus, although PKI␥ is necessary for nuclear export of PKA catalytic subunits, termination of PKA signaling does not depend on nuclear export. Instead, binding and inactivation of PKA by PKI␥ in the nucleus is sufficient to inhibit signaling and expression of immediate-early genes.

DISCUSSION
This study led to two major conclusions. First, endogenous PKI␥ is necessary and sufficient for termination of PKA signaling induced by either PTH or ␤-adrenergic agonists. Thus, PKI␥ knockdown by either siRNA or antisense transfection substantially delays termination of nuclear PKA activity, phosphorylation of transcription factors, and expression of immediate-early genes. The second major conclusion is that PKA catalytic subunit export from the nucleus depends on endogenous PKI␥ but export is not required for termination of PKA signaling and gene expression. Thus, binding and inactivation of PKA is sufficient for PKI␥ to regulate PKA signaling in the nucleus. These findings were the first in any cell type showing that endogenous PKI regulates PKA signaling.
The delayed termination of PKA signaling, transcription factor phosphorylation, and immediate-early gene expression because of PKI␥ knockdown demonstrates the important role of endogenous PKI␥ in termination of these processes. PKI␥ likely acts together with other more downstream mechanisms to maintain low levels of PKA-induced gene expression for longer time periods. For example, CREB dephosphorylation has been implicated in termination of PKA-induced gene expression in fibroblasts; however, it is controversial whether CREB dephosphorylation is primarily due to protein phosphatase-1 or to protein phosphatase-2A (reviewed in Ref. 3). An alternative mechanism for removal of phosphorylated CREB is ubiquitinmediated proteasomal degradation (14). Such degradation may be especially relevant because activation of PKA signaling by PTH up-regulates proteasome activity (48). Our study confirmed that loss of phosphorylated CREB occurs rapidly in cells expressing wild type levels of PKI␥, although we did not investigate whether this is because of dephosphorylation or proteasomal degradation. Expression of inhibitory transcription fac-tors, such as inducible cAMP early repressor, is another downstream mechanism that may act together with PKI␥ to maintain low levels of PKA-induced gene expression (15). In this regard, PTH increases inducible cAMP early repressor (ICER) expression and ICER overexpression decreases stimulation of gene expression by PTH (49). Finally, histone deacetylation or co-activator methylation can also terminate PKAinduced gene expression (16,17). It has recently been shown that histone deacetylase and protein phosphatase-1 form a complex that coordinately regulates CREB phosphorylation and histone acetylation (13). It is therefore interesting to speculate that PKI␥ may also form complexes with molecules responsible for downstream mechanisms in order to coordinately regulate PKA-induced gene expression.
We also found that PKA catalytic subunit export from the nucleus depends on endogenous PKI␥. Moreover, inhibition of nuclear export signal activity blocks nuclear export of both PKI␥ and PKA catalytic subunits despite the fact that the catalytic subunits do not contain nuclear export signal motifs (26). These results are consistent with the model that binding to PKA catalytic subunits exposes the nuclear export signal on PKI␥, thereby inducing co-transport of the catalytic subunit-PKI␥ complex out of the nucleus (25,26). We also found that this nuclear export is not required for termination of PKA signaling and gene expression. Instead, it is likely that PKI␥mediated nuclear export allows PKA catalytic subunits to reassociate with the regulatory subunits in the cytoplasm in preparation for subsequent rounds of cAMP signaling.
Regulation of PKA signaling and gene expression by PKI␥ may have important physiological and clinical implications. For example, primary response genes induced by PKA signaling are responsible for both the catabolic and anabolic effects of PTH and ␤-adrenergic agonists on bone turnover (9,50,51). PTH dosing regimens are being developed to favor the anabolic effects of PTH over its catabolic effects as therapies for diseases that cause bone loss (52). Moreover, ␤-adrenergic antagonists increase bone mass in mice, and it has therefore been proposed that they may also be useful therapies for patients with bone loss (46). Thus, understanding and ultimately being able to manipulate the complex pathways that regulate the balance between the anabolic and catabolic responses induced by PTH and ␤-adrenergic signaling would have important clinical implications. PKI␥ therefore represents a potentially important target for understanding and manipulating this balance between anabolic and catabolic responses to PTH and ␤-adrenergic signaling.
In summary, we found that endogenous levels of PKI␥ are required for termination of both PKA signaling and immediateearly gene expression. We also found that nuclear export of PKA depends on PKI␥ but is not required for termination of either PKA signaling or gene expression. Because this study focused on the role of PKI in osteoblasts, it is particularly relevant for understanding the complex balance between the catabolic and anabolic effects of PKA signaling on bone turn-FIG. 6. PKA catalytic subunit export from the nucleus depends on endogenous PKI␥. PKI␥ knockdown by siRNA (panels 1 and 2) blocks the export of PKA catalytic subunits C␣, C␤, and C␥ back to the cytosol (panels 3-5) that occurs following nuclear translocation induced by PTH. Western blots were performed on nuclear and cytosolic fractions prepared from replicate cultures in the experiment described in Fig. 2. Identity and purity of the fractions were confirmed by assessing CREB as a nuclear marker (panels 6 and 7) and AKT as a cytosolic marker (panels 8 and 9) in aliquots representing equal cell numbers.
FIG. 7. PKA catalytic subunit export from the nucleus is not required for termination of PKA signaling. Inhibition of PKI␥ export back to the cytosol by leptomycin B (LMB) (panel 1) blocks PKA catalytic subunit export back to the cytosol (panels 2-4) (A) but does not alter termination of nuclear PKA activity (B), CREB phosphorylation (C), or immediate-early gene expression (D) following stimulation with PTH. ROS 17/2.8 cells were incubated for the indicated times with 100 nM PTH in the presence or absence of 10 ng/ml LMB. All cultures also received a mixture of vehicle controls, such that all contained 1 M acetic acid and 0.2% ethanol. A, PKI␥ and PKA levels in nuclear fractions were assessed by Western blotting. Identity and purity of the nuclear fractions were confirmed by assessing CREB as a nuclear marker (panel 5) and AKT as a cytosolic marker (panel 6) in aliquots representing equal cell numbers. B, nuclear PKA activity was assessed biochemically. There is no significant difference (p Ͼ0.28) between the two groups at any time point. C, transcription factor phosphorylation was assessed by Western blotting. D, immediate-early gene expression was assessed by RT-PCR. C and D, equivalent loading is documented by examining levels of total CREB protein and actin mRNA, respectively. over described above. However, regulation of gene expression by PKA signaling controls diverse cellular processes in many cell types (10). It is therefore likely that endogenous levels of PKI have important effects on many of these processes.