Scaffolding of adenylyl cyclase by A Kinase Anchoring Proteins

Adenylyl cyclase (AC) converts ATP into cAMP, which activates protein kinase A (PKA). Activation of PKA leads to the phosphorylation of specific substrates. The mechanism of specificity of PKA phosphorylation baffled researchers for many years. The discovery of A Kinase Anchoring Proteins (AKAPs) has helped to unravel this mystery. AKAPs function to target PKA to specific regions within the cell. They also anchor other enzymes, receptors, or channels leading to tightly regulated signaling modules. Several studies have suggested an important role for activated PKA in these complexes, including the AKAPs yotiao and muscle AKAP (mAKAP). Yotiao, a plasma membrane AKAP, anchors PP1, NMDA receptors, IP3 receptors, and heart potassium channel subunit KCNQI. PKA phosphorylation of NMDA receptors as well as KCNQI leads to increased channel activity. Patients with mutations in KCNQI or yotiao that cause loss of targeting of KCNQI develop long QT syndrome, which can be fatal. mAKAP anchors several CAMP/PKA-regulated pathways to the nuclear envelope in cardiac myocytes. The necessity of activated PKA in these complexes led to the hypothesis that AC is also anchored. The results indicate that AC does associate with yotiao in brain and heart, specifically with AC types I-III, and IX. Co-expression of AC II or III with yotiao leads to inhibition of each isoform's activity. Binding assays revealed that yotiao binds to the N-terminus of AC II and that this region can reverse the inhibition of AC II, but not AC III, indicating unique binding sites on yotiao. AC II binds directly to as 808-957 of yotiao. Y808-957 acts as a dominant negative as the addition of it to rat brain membranes results in a ∼40% increase in AC activity. Additionally, AC was also found to associate with mAKAP in heart, specifically with AC types II and V. The binding site of AC was mapped to 275-340 of mAKAP, while mAKAP binds to the soluble domains of AC V as a complex. These results indicate that interactions between AC and AKAPs are specific and that AC plays an important role in AKAP-targeted signaling. ^

The regulation of mTORC2 kinase activity and complex integrity

Growth factor signaling promotes anabolic processes via activation of the PI3K-Akt kinase cascade. Deregulation of the growth factor-dependent PI3K-Akt pathway was implicated in tumorigenesis. Akt is an essential serine/threonine protein kinase that controls multiple physiological functions such as cell growth, proliferation, and survival to maintain cellular homeostasis. Recently, the mammalian Target of Rapamycin Complex 2 (mTORC2) was identified as the main Akt Ser-473 kinase, and Ser-473 phosphorylation is required for Akt hyperactivation. However, the detailed mechanism of mTORC2 regulation in response to growth factor stimulation or cellular stresses is not well understood. In the first project, we studied the regulation of the mTORC2-Akt signaling under ER stress. We identified the inactivation of mTORC2 by glycogen synthase kinase-3β (GSK-3β). Under ER stress, the essential mTORC2 component, rictor, is phosphorylated by GSK-3β at Ser-1235. This phosphorylation event results in the inhibition of mTORC2 kinase activity by interrupting Akt binding to mTORC2. Blocking rictor Ser-1235 phosphorylation can attenuate the negative impacts of GSK-3β on mTORC2/Akt signaling and tumor growth. Thus, our work demonstrated that GSK-3β-mediated rictor Ser-1235 phosphorylation in response to ER stress interferes with Akt signaling by inhibiting mTORC2 kinase activity. In the second project, I investigated the regulation of the mTORC2 integrity. We found that basal mTOR kinase activity depends on ATP level, which is tightly regulated by cell metabolism. The ATP-sensitive mTOR kinase is required for SIN1 protein phosphorylation and stabilization. SIN1 is an indispensable subunit of mTORC2 and is required for the complex assembly and mTORC2 kinase activity. Our findings reveal that mTOR-mediated phosphorylation of SIN1 is critical for maintaining complex integrity by preventing SIN1 from lysosomal degradation. In sum, our findings verify two distinct mTORC2 regulatory mechanisms via its components rictor and SIN1. First, GSK-3β-mediated rictor Ser-1235 phosphorylation results in mTORC2 inactivation by interfering its substrate binding ability. Second, mTOR-mediated Ser-260 phosphorylation of SIN1 preserves its complex integrity. Thus, these two projects provide novel insights into the regulation of mTORC2.

Regulation of myogenesis by growth signals -growth factors, oncogenes and protein kinases

Establishment of a myogenic phenotype involves antagonism between cell proliferation and differentiation. The recent identification of the MyoD family of muscle-specific transcription factors provides opportunities to dissect at the molecular level the mechanisms through which defined cell type-specific transcription factors respond to environmental cues and regulate differentiation programs. This project is aimed at elucidation of the molecular mechanism whereby growth factors repress myogenesis. Initial studies demonstrated that nuclear oncogenes such as c-fos, junB and c-jun are immediate early genes that respond to serum and TGF-$\beta$. Using the muscle creatine kinase (MCK) enhancer linked to the reporter gene CAT as a marker for differentiation, we showed that transcriptional function of myogenin can be disrupted in the presence of c-Fos, JunB and cjun. In contrast, JunD, which shares DNA-binding specificity with JunB and c-Jun but is expressed constitutively in muscle cells, failed to show the inhibition. The repression by Fos and Jun is targeted at KE-2 motif, the same sequence that mediates myogenin-dependent activation and muscle-specific transactivation. Deletion analysis indicated that the transactivation domain of c-Jun at the N-terminus is responsible for the repression. Considering that myogenin is a phosphoprotein and cAMP and TPA are able to regulate myogenesis, we examined whether constitutively active protein kinase C (PKC) and protein kinase A (PKA) could substitute for exogenous growth factors and prevent transcription activation by myogenin. Indeed, the basic region of myogenin is phosphorylated by PKC at a threonine that is conserved in all members of the MyoD family. Phosphorylation at this site attenuates DNA binding activity of myogenin. Protein kinase A can also phosphorylate myogenin in a region adjacent to the DNA binding domain. However, phosphorylation at this site is insufficient to abrogate myogenin's DNA binding capacity, suggesting that PKA and PKC may affect myogenin transcriptional activity through different mechanisms. These findings provide insight into the mechanisms through which growth factor signals negatively regulate the muscle differentiation program and contribute to an understanding of signal transducing pathways between the cell membrane and nucleus. ^

Regulation of v-Mos kinase activity by phosphorylation

In this thesis, I investigated the effect of cylic AMP-dependent protein kinase (PKA) on v-Mos kinase activity. Increase in PKA activity in vivo brought about either by forskolin treatment or by overexpression of the PKA catalytic subunit resulted in a significant inhibition of v-Mos kinase activity. The purified PKA catalytic subunit was able to phosphorylate recombinant p37$\rm\sp{v-mos}$ in vitro, suggesting that the mechanism of in vivo inhibition of v-Mos kinase involves direct phosphorylation by PKA. Ser-263 was identified as a residue that is normally phosphorylated at a very low level but whose phosphorylation is dramatically increased upon forskolin treatment. Consistent with the inhibitory role of Ser-263 phosphorylation, the Ala-263 mutant of v-Mos was not inhibited by forskolin treatment. Based on our results, we propose that the known inhibitory role of PKA in the initiation of oocyte maturation could be explained at least in part by its inhibition of Mos kinase.^ Combining tryptic phosphopeptide two-dimensional mapping analysis and in vitro mutagenesis studies, I identified Ser-56 as the major in vivo phosphorylation site on v-Mos. I studied the interrelationship between Ser-34 and Ser-56 phosphorylation in regulating v-Mos function. After site-directed mutagenesis to substitute serine residues with alanine or glutamic acid in different combinations to mimick unphosphorylated and phosphorylated serines respectively, various v-Mos mutants were expressed in COS-1 cells. As expected, Ala-34 mutant of v-Mos had very low (less 5% of wild type) kinase activity. The Ala-56 mutant had kinase activity 50% that of wild type. Surprisingly, the Ala-34 Ala-56 double mutant and the Ala-56 mutant exhibited identical kinase activity. On the other hand, Ala-34 Glu-56 double mutant had reduced kinase activity comparable to Ala-34 mutant. These results suggest that the phosphorylation at Ser-56 may serve to inhibit the activation of newly synthesized Mos protein. As predicted from Xenopus c-Mos studies, Glu-34 mutant of v-Mos was highly active (125% that of wild type). Interestingly, consistant with the model involving an inhibitory role of Ser-56 phosphorylation, the Glu-34 Glu-56 double mutant was totally inactive as a kinase. Moreover in my experiments, there was a perfect correlation between the level of v-Mos kinase activity of various mutants and their transforming activity. The latter is dependent upon MEK1 phosphorylation/ activation in v-mos transformed cells. Residues corresponding to both v-Mos Ser-34 and Ser-56 are evolutionarily conserved in c-Mos. Therefore, the cytostatic factor function of c-Mos may be regulated in the same manner as v-Mos kinase activity.^ It has been known that v-mos transforms cells by affecting G1 phase progression of the cell cycle. Here I showed that mos induces cyclin D1 expression in mos transformed NIH 3T3 cells and NRK 6m2 cells, and this induced level was found to be unaffected by serum starvation. Consequently, cyclin D1-Cdk4 and cyclin E-Cdk2 activities increase, and retinoblastoma protein is hyperphosphorylated. Based on studies from several laboratories, these findings suggest that increased amount of cyclin D1-Cdk4 complexes ties up the limited amount of cyclin E-Cdk2 inhibitors (e.g. p27), causing the activation of cyclin E-Cdk2. My results indicate that activation of key cell cycle regulators of G1 phase may be important for cellular transformation by mos. (Abstract shortened by UMI.) ^

The Shc adaptor protein and its role on Her -2 /neu oncogenesis

Shc proteins are implicated in coupling receptor tyrosine kinases to the mitogen-activated protein kinase (MAPK) pathway by recruiting Grb2/SOS to the plasma membrane. To better understand the role of Shc in oncogenesis brought about by point mutation activated neu (p185*), we transfected a Shc mutant (ShcΔCH1), which lacks the Grb2 binding site Y317 by deletion of collagen-homology domain 1, into p185*-transformed NIH3T3 cells. The cellular transformation phenotypes were found to be largely suppressed by expression of ShcΔCH1. This study indicates that Shc plays a critical role in mediating the oncogenical signals of p185*. Although ShcΔCH1 still retained another Grb2 binding site (Y239/240), we did not detect its physical association with Grb2. We also found that ShcΔCH1 could associate with p185*; however, this association did not interfere with the endogenous Shc-p185* interaction or the Shc-Grb2 interaction. In addition, p185*-mediated MAPK/Elk activation, PI3-K activation and Src activation likewise was not inhibited by ShcΔCH1 expression. Taken together, our current study clearly indicates that ShcΔCH1 suppresses the p185*-induced transformation, and that this suppression is mediated through a MAPK-independent and possibly PI3-K, Src-independent pathway. These results suggest that Shc may be involved in other unidentified signal pathways which are critical for p185*-induced cellular transformation besides the three pathways that we have studied. ^

Molecular mechanism of the relaxin signaling pathways: Activation of cAMP, PI3K, and PKCzeta

Relaxin is a polypeptide hormone that has diverse effects on reproductive and non-reproductive tissues. Relaxin activates the G-protein coupled receptors, LGR7 and LRG8. Early studies described increased cAMP and protein kinase A activity upon relaxin treatment, but cAMP accumulation alone could not account for all of the relaxin-mediated effects. We utilized the human monocyte cell line THP-1 to study the mechanism of relaxin-stimulated CAMP production. ^ Relaxin treatment in THP-1 cells produces a biphasic time course in cAMP accumulation, where the first peak appears as early as 1–2 minutes with a second peak at 10–20 minutes. Selective inhibitors for phosphoinositide 3-kinase (P13K), such as wortmannin and LY294002, show a dose-dependent inhibition of relaxin-stimulated cAMP accumulation, specific for the second peak of the relaxin time course. Neither the effects of relaxin nor the inhibition of relaxin by LY294002 is mediated by the activity of phosphodiesterases. Furthermore, LY294002 blocks upregulation of vascular endothelial growth factor transcript levels by relaxin. ^ To further delineate relaxin signaling pathways, we searched for downstream targets of PI3K that could activate adenylyl cyclase (AC). Protein kinase C ζ (PKCζ) was a prime candidate because it activates types II and V AC. Chelerythrine chloride (a general PKC inhibitor) inhibits relaxin-induced cAMP production to the same degree as LY294002 (∼40%). Relaxin stimulates PKCζ translocation to the plasma membrane in THP-1, MCF-7, PHM1-31, and MMC cells, as shown by immunocytochemistry. PKCζ translocation is P13K-dependent and independent of cAMP production. Antisense PKCζ oligodeoxynucleotides (PKCζ-ODNs) deplete both PKCζ transcript and protein levels in THP-1 cells. PKCζ-ODNs abolish relaxin-mediated PKCζ translocation and inhibit relaxin stimulation of cAMP by 40%, as compared to mock and random ODN controls. Treatment with LY294002 in the presence of PKCζ-ODNs results in little further inhibition. Taken together, we present a novel role for PI3K and PKCζ in relaxin stimulation of cAMP and provide the first example of the PKCζ regulation of AC in an endogenous system. Furthermore, we have identified higher order complexes of AC isoforms and PKA anchoring proteins in attempts to explain the differential coupling of relaxin to cAMP and PI3K-signaling pathways in various cell types. ^

Cell adhesion and the integrin-linked kinase regulate the LEF-1 and β-catenin signaling pathways

The integrin-linked kinase (ILK) is an ankyrin repeat containing serine-threonine protein kinase that can interact directly with the cytoplasmic domains of the β1 and β3 integrin subunits and whose kinase activity is modulated by cell–extracellular matrix interactions. Overexpression of constitutively active ILK results in loss of cell–cell adhesion, anchorage-independent growth, and tumorigenicity in nude mice. We now show that modest overexpression of ILK in intestinal epithelial cells as well as in mammary epithelial cells results in an invasive phenotype concomitant with a down-regulation of E-cadherin expression, translocation of β-catenin to the nucleus, formation of a complex between β-catenin and the high mobility group transcription factor, LEF-1, and transcriptional activation by this LEF-1/β-catenin complex. We also find that LEF-1 protein expression is rapidly modulated by cell detachment from the extracellular matrix, and that LEF-1 protein levels are constitutively up-regulated at ILK overexpression. These effects are specific for ILK, because transformation by activated H-ras or v-src oncogenes do not result in the activation of LEF-1/β-catenin. The results demonstrate that the oncogenic properties of ILK involve activation of the LEF-1/β-catenin signaling pathway, and also suggest ILK-mediated cross-talk between cell–matrix interactions and cell–cell adhesion as well as components of the Wnt signaling pathway.

Arachidonic acid mediates angiotensin II effects on p21ras in renal proximal tubular cells via the tyrosine kinase-Shc-Grb2-Sos pathway

In kidney epithelial cells, an angiotensin II (Ang II) type 2 receptor subtype (AT2) is linked to a membrane-associated phospholipase A2 (PLA2) and the mitogen-activated protein kinase (MAPK) superfamily. However, the intervening steps in this linkage have not been determined. The aim of this study was to determine whether arachidonic acid mediates Ang II’s effect on p21ras and if so, to ascertain the signaling mechanism(s). We observed that Ang II activated p21ras and that mepacrine, a phospholipase A2 inhibitor, blocked this effect. This activation was also inhibited by PD123319, an AT2 receptor antagonist but not by losartan, an AT1 receptor antagonist. Furthermore, Ang II caused rapid tyrosine phosphorylation of Shc and its association with Grb2. Arachidonic acid and linoleic acid mimicked Ang II-induced tyrosine phosphorylation of Shc and activation of p21ras. Moreover, Ang II and arachidonic acid induced an association between p21ras and Shc. We demonstrate that arachidonic acid mediates linkage of a G protein-coupled receptor to p21ras via Shc tyrosine phosphorylation and association with Grb2/Sos. These observations have important implications for other G protein-coupled receptors linked to a variety of phospholipases.

Dissociation of cytokine-induced phosphorylation of Bad and activation of PKB/akt: Involvement of MEK upstream of Bad phosphorylation

The phosphatidylinositol 3-kinase (PI3K)-signaling pathway has emerged as an important component of cytokine-mediated survival of hemopoietic cells. Recently, the protein kinase PKB/akt (referred to here as PKB) has been identified as a downstream target of PI3K necessary for survival. PKB has also been implicated in the phosphorylation of Bad, potentially linking the survival effects of cytokines with the Bcl-2 family. We have shown that granulocyte/macrophage colony-stimulating factor (GM-CSF) maintains survival in the absence of PI3K activity, and we now show that when PKB activation is also completely blocked, GM-CSF is still able to stimulate phosphorylation of Bad. Interleukin 3 (IL-3), on the other hand, requires PI3K for survival, and blocking PI3K partially inhibited Bad phosphorylation. IL-4, unique among the cytokines in that it lacks the ability to activate the p21ras–mitogen-activated protein kinase (MAPK) cascade, was found to activate PKB and promote cell survival, but it did not stimulate Bad phosphorylation. Finally, although our data suggest that the MAPK pathway is not required for inhibition of apoptosis, we provide evidence that phosphorylation of Bad may be occurring via a MAPK/ERK kinase (MEK)-dependent pathway. Together, these results demonstrate that although PI3K may contribute to phosphorylation of Bad in some instances, there is at least one other PI3K-independent pathway involved, possibly via activation of MEK. Our data also suggest that although phosphorylation of Bad may be one means by which cytokines can inhibit apoptosis, it may be neither sufficient nor necessary for the survival effect.

Requirement of GM2 ganglioside activator for phospholipase D activation

Sequence analysis of a heat-stable protein necessary for the activation of ADP ribosylation factor-dependent phospholipase D (PLD) reveals that this protein has a structure highly homologous to the previously known GM2 ganglioside activator whose deficiency results in the AB-variant of GM2 gangliosidosis. The heat-stable activator protein indeed has the capacity to enhance enzymatic conversion of GM2 to GM3 ganglioside that is catalyzed by β-hexosaminidase A. Inversely, GM2 ganglioside activator purified separately from tissues as described earlier [Conzelmann, E. & Sandhoff, K. (1987) Methods Enzymol. 138, 792–815] stimulates ADP ribosylation factor-dependent PLD in a dose-dependent manner. At higher concentrations of ammonium sulfate, the PLD activator protein apparently substitutes for protein kinase C and phosphatidylinositol 4,5-bisphosphate, both of which are known as effective stimulators of the PLD reaction. The mechanism of action of the heat-stable PLD activator protein remains unknown.

Hepatitis B virus HBx protein sensitizes cells to apoptotic killing by tumor necrosis factor α

Persistent infection with hepatitis B virus (HBV) is a leading cause of human liver disease and is strongly associated with hepatocellular carcinoma, one of the most prevalent forms of human cancer. Apoptosis (programmed cell death) is an important mediator of chronic liver disease caused by HBV infection. It is demonstrated that the HBV HBx protein acutely sensitizes cells to apoptotic killing when expressed during viral replication in cultured cells and in transfected cells independently of other HBV genes. Cells that were resistant to apoptotic killing by high doses of tumor necrosis factor α (TNFα), a cytokine associated with liver damage during HBV infection, were made sensitive to very low doses of TNFα by HBx. HBx induced apoptosis by prolonged stimulation of N-Myc and the stress-mediated mitogen-activated-protein kinase kinase 1 (MEKK1) pathway but not by up-regulating TNF receptors. Cell killing was blocked by inhibiting HBx stimulation of N-Myc or mitogen-activated-protein kinase kinase 1 using dominant-interfering forms or by retargeting HBx from the cytoplasm to the nucleus, which prevents HBx activation of cytoplasmic signal transduction cascades. Treatment of cells with a mitogenic growth factor produced by many virus-induced tumors impaired induction of apoptosis by HBx and TNFα. These results indicate that HBx might be involved in HBV pathogenesis (liver disease) during virus infection and that enhanced apoptotic killing by HBx and TNFα might select for neoplastic hepatocytes that survive by synthesizing mitogenic growth factors.

Independent regulation of JNK/p38 mitogen-activated protein kinases by metabolic oxidative stress in the liver

The stress-activated protein kinases JNK and p38 mediate increased gene expression and are activated by environmental stresses and proinflammatory cytokines. Using an in vivo model in which oxidative stress is generated in the liver by intracellular metabolism, rapid protein–DNA complex formation on stress-activated AP-1 target genes was observed. Analysis of the induced binding complexes indicates that c-fos, c-jun, and ATF-2 were present, but also two additional jun family members, JunB and JunD. Activation of JNK precedes increased AP-1 DNA binding. Furthermore, JunB was shown to be a substrate for JNK, and phosphorylation requires the N-terminal activation domain. Unexpectedly, p38 activity was found to be constitutively active in the liver and was down-regulated through selective dephosphorylation following oxidative stress. One potential mechanism for p38 dephosphorylation is the rapid stress-induced activation of the phosphatase MKP-1, which has high affinity for phosphorylated p38 as a substrate. These data demonstrate that there are mechanisms for independent regulation of the JNK and p38 mitogen-activated protein kinase signal transduction pathways after metabolic oxidative stress in the liver.

Preferential activation of the p46 isoform of JNK/SAPK in mouse macrophages by TNFα

A pleiotropic cytokine, tumor necrosis factor-α (TNFα), regulates the expression of multiple macrophage gene products and thus contributes a key role in host defense. In this study, we have investigated the specificity and mechanism of activation of members of the c-Jun-NH2-terminal kinase/stress-activated protein kinase (JNK/SAPK) subfamily of mitogen-activated protein kinases (MAPKs) in mouse macrophages in response to stimulation with TNFα. Exposure of macrophages to TNFα stimulated a preferential increase in catalytic activity of the p46 JNK/SAPK isoform compared with the p54 JNK/SAPK isoform as determined by: (i) separation of p46 and p54 JNK/SAPKs by anion exchange liquid chromatography and (ii) selective immunodepletion of the p46 JNK/SAPK from macrophage lysates. To investigate the level of regulation of p46 JNK/SAPK activation, we determined the ability of MKK4/SEK1/JNKK, an upstream regulator of JNK/SAPKs, to phosphorylate recombinant kinase-inactive p46 and p54 JNK/SAPKs. Endogenous MKK4 was able to transphosphorylate both isoforms. In addition, both the p46 and p54 JNK/SAPK isoforms were phosphorylated on their TPY motif in response to TNFα stimulation as reflected by immunoblotting with a phospho-specific antibody that recognizes both kinases. Collectively, these results suggest that the level of control of p46 JNK/SAPK activation is distal not only to MKK4 but also to the p54 JNK/SAPK. Preferential isoform activation within the JNK/SAPK subfamily of MAPKs may be an important mechanism through which TNFα regulates macrophage phenotypic heterogeneity and differentiation.

Modes of action of aspirin-like drugs: Salicylates inhibit Erk activation and integrin-dependent neutrophil adhesion

The anti-inflammatory effects of high-dose salicylates are well recognized, incompletely understood and unlikely due entirely to cyclooxygenase (COX) inhibition. We have previously reported a role for activation of the kinase Erk in CD11b/CD18 integrin-dependent adhesiveness of human neutrophils, a critical step in inflammation. We now report the effects of salicylates on neutrophil Erk and adhesion. Exposure of neutrophils to aspirin or sodium salicylate (poor COX inhibitor) inhibited Erk activity and adhesiveness of formylmethionyl-leucyl-phenylalanine- and arachidonic acid-stimulated neutrophils, consistent with anti-inflammation but not COX inhibition (IC50s = 1–8 mM). In contrast, indomethacin blocked neither Erk nor adhesion. Inhibition of Mek (proximal activator of Erk) also blocked stimulation of Erk and adhesion by formylmethionyl-leucyl-phenylalanineand arachidonic acid. Salicylate inhibition of Erk was independent of protein kinase A activation and generation of extracellular adenosine. These data are consistent with a role for Erk in stimulated neutrophil adhesion, and suggest that anti-inflammatory effects of salicylates may be mediated via inhibition of Erk signaling required for integrin-mediated responses.

Odorants induce the phosphorylation of the cAMP response element binding protein in olfactory receptor neurons

Although odorants are known to activate olfactory receptor neurons through cAMP, the long-term effects of odorant detection are not known. Our recent findings indicate that there is also a delayed and sustained cAMP response, with kinetics sufficient to mediate long-term cellular responses. This cAMP response is mediated by cGMP through activation of adenylyl cyclase by protein kinase G (PKG). Therefore, we investigated the ability of odorants to regulate gene expression in rat olfactory epithelium. The cAMP-responsive binding protein (CREB) is a well-characterized transcription factor regulated by cAMP. We examined CREB activity in rat olfactory epithelium and olfactory receptor neurons (ORNs) after stimulation with odorants. Odorants increased levels of phosphorylated CREB in olfactory epithelium in vivo, and this increase was localized to ORNs in vitro. Incubation with 8-bromo-cGMP or sodium nitroprusside, a guanylyl cyclase activator, also increased phosphorylated CREB. In vitro, cAMP-dependent protein kinase phosphorylated CREB. In contrast, PKG failed to phosphorylate CREB directly in vitro. Our results demonstrate that the delayed odorant-induced cAMP signal activates CREB, which in turn may modulate gene expression in ORNs. In addition, cGMP indirectly affects CREB activation. This effect of cGMP on CREB activity through cAMP provides another mechanism for the modulation of CREB.