Complex roles of cAMP–PKA–CREB signaling in cancer

Abstract

Background

Cell growth is tightly regulated by a variety of signal transduction pathways. Abnormal activation or inhibition of signal transduction pathways drives tumorigenesis. Posttranslational protein modifications, such as phosphorylation, ubiquitination, methylation and acetylation, are one of the important mechanisms that regulate cell signaling. Protein phosphorylation and dephosphorylation are regulated by various protein kinases and phosphatases, respectively. Abnormal or uncontrolled activation of protein kinases or phosphatases is very common in tumors, poising them as important targets for molecular targeted cancer therapeutics. As an inhibitor of the oncogenic kinase BCR–ABL, imatinib is the first kinase-targeted anticancer drug that has been successfully applied in the treatment of chronic myeloid leukemia [1]. Later on, inhibitors of protein kianses such as EGFR, ErbB2, MAPK, VEGFR and mTOR have been widely used in the treatment of a variety of common malignant tumors [2–5].

Cell signaling is often initiated by first messengers such as growth factors, hormones and ions, which trigger a series of signal transduction cascades via membrane receptors or intracellular receptors. This process involves multiple feedback mechanisms as well as a number of intracellular chemicals, known as the second messengers, such as cyclic adenosine monophosphate (cAMP), cGMP, calcium ions and so on [6]. In general, the production and distribution of these second messengers also need precise regulation. Abnormal production and distribution of these second messengers may contribute to carcinogenesis and tumor progression. The second messenger theory was first proposed by E. W. Sutherland in 1965. As a second messenger, cAMP is responsible for activation of the protein kinase A (PKA), exchange proteins directly activated by cAMP (Epac) and ion gated channel protein [7]. As one of the target proteins of PKA, cAMP response element binding protein (CREB) is an important transcription factor that regulates the expression of several genes including oncogenes c-Jun and cyclin D1 [8]. Notably, cAMP–PKA–CREB signaling has both tumor-suppressive and tumot-promoting effects on cancer, depending on the tumor types and context.

The generation and degradation of cAMP

cAMP exists extensively in cells. Many hormones, neurotransmitters and other signaling molecules use it as intracellular second messenger. Therefore, cAMP can directly regulate various biological processes or behaviors of cells, including cell metabolism, ion channel activation, gene expression, cell growth, differention and apoptosis [9]. The generation of cAMP is regulated in a G-protein-dependent or G-protein-independent manner. After extracellular ligands, such as PGE2, GLP-1 and β2 receptor agonists, bind to G-protein coupled receptors (GPCRs), Gα subunits are separated from Gβ and Gγ subunits, and then activate adenylyl cyclases (ACs), leading to the conversion of ATP into cAMP (Fig. 1) [10–12]. In addition, bicarbonate (HCO₃⁻) and calcium ions (Ca²⁺) induce cAMP synthesis by activating the soluble adenylyl cyclase (sAC) independent of G-proteins [13, 14]. In contrast, phosphodiesterases (PDEs) are responsible for the degradation of cAMP. So far, at least 22 PDEs have been identified [15]. The concentration of intracellular cAMP depends on the relative balance between adenylyl cyclases and phosphodiesterases.

Open in a separate window

Fig. 1

cAMP signaling pathway. The generation of cAMP is regulated in a G-protein dependent manner by AC or G-protein independent manner by sAC. The degradation of cAMP is regulated by PDEs. cAMP can regulate multiple signaling pathways including ion channes, Epac, and PKA. R the regulatory subunit of PKA, C the catalytic subunit of PKA

The molecular targets and roles of cAMP in cancer cell growth, migration and metabolism

As shown in Fig. 1, the major targets of cAMP include PKA, Epac1 and Epac2, and nucleotide-gated ion channels. The cAMP binding domain of Epac can bind a cAMP molecule, resulting in conformational changes of the protein thereby exposing the active site in catalytic domain. As a guanosine exchange factor (GEF), Epac can promote Ras family proteins Rap1 and Rap2 to release GDP and bind GTP, leading to the activation of Rap1/2 [16]. Epac regulates multiple cell processes mainly through Rap-GEF dependent and Rap-GEF independent pathways, including migration and focal adhesion formation, morphogenesis, cardiac output, neurotransmitter release, glucose homeostasis, exocytosis, cell proliferation, cell differentiation and cell survival [16]. cAMP can also bind to some nucleotide-gated ion channels and regulate their functions, such as cyclic nucleotide gated ion channels (CNG) and hyperpolarization-activated cyclic nucleotide-gated channel (HCN) [6]. These channels are non-selective cation channels that can conduct calcium, sodium and potassium ions and change the cell membrane potential. Below, we shall focus on PKA as a classical target of cAMP.

PKA is a tetramer enzyme consisting of two regulatory subunits (R) and two catalytic subunits (C) (Fig. 1). In mammals, there are four types of regulatory subunits, namely RIα, RIβ, RIIα and RIIβ. According to the different regulatory subunits, PKA can be classified into PKA type I (RIα₂C₂, RIβ₂C₂) and PKA type II (RIIα₂C₂, RIIβ₂C₂) [17]. The binding to two regulatory subunits of the inactive PKA tetramer by cAMP results in the release and activation of the catalytic subunits, phosphorylation of the serine and threonine residues in substrate proteins [18]. RIα and RIIα are widely distributed in different cells, while RIβ mainly exists in the brain, testis and B lymphocytes, and RIIβ is expresses in the brain, fat, and some endocrine tissues [19]. The localization of the tetramer enzyme is also different. PKA type I are generally cytoplasmic, while PKA type II are specifically anchored in subcellular structures and compartments [20]. PKA anchored proteins (AKAPs) can bind to cytoskeleton proteins or organelles and to PKA regulatory subunits, thereby playing an important role in the distribution of PKA in cells and enabling PKA to park and concentrate key targets [21, 22]. PKA type II is generally expressed in normal non-proliferating tissues and cells in growth retardation phase. PKA type I expression can be temporarily induced by physiological stimulation in cell proliferating phase and stimulated by certain growth factors and receptors, such as EGFR, TNF-α, and ErbB-2. Moreover, PKA type I is overexpressed in a variety of primary tumors [23].

There are three types of PKA catalytic subunits, including Cα, Cβ and Cγ. Cα and Cβ are two major types and have multiple splice variants. This subtype diversity is an important mechanism for achieving the specificity of PKA signaling [24]. Once activated, PKA phosphorylates its substrates, such as CREB, Raf, Bad and GSK3 [25–27], and then regulates gene expression, cell survival and migration. Recent study indicates that PKA is an actomyosin contractility-regulated effector of cellular mechanotransduction and a regulator of mechanically guided cell migration [28]. PKA also phosphorylates CDC42 interacting protein 4 (CIP4), a coordinator of membrane deformation and actin polymerization, and promotes cancer cell invasion and metastasis [29]. Actually, the promotion of cancer cell invasion and metastasis by PKA may involve multiple PKA substrates. The focal adhesion kinase is another PKA target that mediates the promotion of cancer metastasis by cAMP [30].

In addition, PKA can phosphorylate and then inactivate the calmodulin-dependent protein kinase kinase-2 (CAMKK2) [31]. CaMKK2 plays important roles in energy homeostasis, insulin signaling and whole-body metabolism [32, 33]. Inhibition of CAMKK2 protects against prostate cancer, hepatocellular carcinoma (HCC) and metabolic disorders induced by a high-fat diet [34]. Moreover, PKA can reprogram lipid metabolism by inhibiting salt-inducible kinases, and then promote pancreatic tumorigenesis [35]. Meanwhile, PKA signaling cooperates with the epigenetic regulators JMJD3 and SIRT1 to activate β-oxidation-promoting genes [36]. There may be more and more metabolism-related targets of PKA to be uncovered. Of note, activated PKA can feedback inhibit adenylyl cyclase (AC5 and AC6) and activate phosphodiesterase (PDE3 and PDE4) thereby lowering the levels of cAMP and maintaing cAMP homeostasis [12, 37].

The CREB family proteins are well-characterized PKA substrates. This family is a group of the basic leucine zipper (bZIP) superfamily transcription factors, including CREB, cAMP responsive element regulatory protein (CREM) and transcriptional activator 1 (ATF1). CREB, CREM and ATF1 can form homodimers or heterodimers [38–40]. Both CREB and ATF-1 proteins are ubiquitously expressed, while CREM is mostly expressed in neuroendocrine tissues [40]. The CREB family proteins contain a kinase inducible domain (KID), two glutamate domains (Q1 and Q2), and a basic leucine zipper domain (bZIP). KID is a 60-amino acid fragment located in the central region of CREB, CREM and ATF1, which contains the PKA phosphorylation site (RRPSY) (Fig. 2) [41].

Open in a separate window

Fig. 2

The domains of CREB family proteins. The CREB family proteins contain a kinase inducible domain (KID), two glutamate domains (Q1 and Q2), and a basic leucine zipper domain (bZIP)

Phosphorylation of CREB Ser¹³³ residue and ATF1 Ser⁶³ residue is critical for the transcriptional activity of CREB and ATF1. Activated ATF1 and CREB can form homodimer or heterodimer, bind to the cAMP response element (CRE) in the promoter region of target genes and initiate genes transcription, thereby regulating cell differentiation, proliferation, apoptosis, metabolism, glucose homeostasis, hematopoiesis, immune response as well as neuronal activities such as memory and learning (Fig. 3) [42–44]. In addition to the major PKA kinase, other kinases such as RSK, MSK, AKT, and CAMKII/IV also directly phosphorylate CREB and ATF1 [45–50].

Open in a separate window

Fig. 3

Cell signaling pathways for activating CREB/ATF1. CREB/ATF1 can be phosphorylated by multiple kinases including Akt, RSK, MSK, PKA, CAMKII and CAMKIV

The roles of cAMP–PKA–CREB signaling pathway in tumors

cAMP–PKA–CREB signaling has paradoxical effects in tumors, acts as a tumor-suppressor or tumor-promoter in different tumor types (Table ​(Table1).1). The role of cAMP–PKA–CREB signaling pathway in liver cancer and other tumors is described below.

Targeting cAMP–PKA pathway for cancer therapy

Given the involvement of cAMP–PKA pathway in tumor progression, targeting this pathway may be a potential strategy to treat cancer. The cAMP analogue 8-Cl-cAMP, also known as tacladesine, can inhibit the growth of colon cancer, breast cancer, lung cancer, fibrosarcoma and leukemia in vitro and in vivo [136, 137]. Due to the different binding affinity with PKA R subunits, 8-Cl-cAMP can down-regulate RIα and induce RIIβ expression [136]. Some researchers have proposed that 8-Cl-cAMP may also exert PKA-independent anti-tumor effects through its metabolite 8-Cl-adenosine and AKT2–PKBβ pathway [138, 139]. Although its anti-tumor mechanisms have not been fully elucidated, 8-Cl-cAMP has been tested in phase II clinical trials in patients with multiple myeloma and in phase I clinical trials in patients with metastatic colorectal cancer, but the results of these trials have not been disclosed (Table ​(Table22).

Since cAMP has tumor suppressive roles in some types of cancer, increasing cAMP levels may inhibit these types of tumors. PDE inhibitors can suppress the depletion of cAMP. Notably, PDE inhibitors have been widely used in clinic to treat cardiovascular, respiratory and psychiatic diseases. The PDE3 inhibitors amrinone and milrinone are used as cardiotonic drugs. PDE4 inhibitor rolipram is a new anti-inflammatory drug for asthma treatment. PDE5 inhibitor sildenafil is a preferred drug for erectile dysfunction treatment [140]. In addition, some flavonoids natural products can also inhibit PDEs [141]. These PDE inhibitors may be repurposed to treat cancers in which cAMP has tumor-suppressive roles. In contrast, PDE activators may be developed to treat cancers that are promoted by cAMP signaling. Recently, a small-molecule compound that allosterically activates PDE4 long isoforms has been developed [142]. This prototypical PDE4 activator compound is able to reduce intracellular cAMP levels and inhibit cyst formation in the kidney [142]. It remains to know whether such kinds of PDE activators may be repurposed to treat some types of cancer.

Antisense oligonucleotide targeting the N-terminal PKA RIα is another choice to interfere with PKA signaling. PKA RIα knockdown leads to compensative expression of RIIβ and then inhibits tumor growth in a wide variety of tumor models [143]. The antisense oligonucleotide GEM231 has been evaluated in two phase I clinical trials (Table ​(Table2).2). The results of the phase I clinical trial of GEM231 in combination with docetaxel showed that the overall incidence of grade 3 adverse reactions, including fatigue, elevated aminotransferase, neutropenia and altered mental status, was 75% in 20 patients with refractory solid tumors during 39 cycles of treatment [144]. No subsequent clinical trial has been reported.

CREB may also be a target for cancer therapy. Currently, there is no specific CREB inhibitor. The small molecule XX-650-23 blocks the interaction between CREB and its co-activator CBP (CREB-binding protein) and then abrogates CREB-responsive gene expression, leading to AML cells apoptosis and cell-cycle arrest [145, 146]. Another feasible strategy is to downregulate CREB expression. Previous study demonstrates that GSKJ4, the histone lysine demethylases JMJD3/UTX inhibitor, can induce CREB degradation and inhibit AML cell growth [123]. Given that JMJD3/UTX have multiple targets, GSKJ4 may also inhibit cancer cell growth by targeting proteins other than CREB. Since CREB may be activated by CAMK in some types of cancer, inhibition of CAMK may be an alternative strategy to suppress CREB [147].

Although there are some PKA inhibitors that have tumor-suppressive effects in preclinical studies, no small-moleculae PKA inhibitors have been tested in clinical trials. It warrants further research to develop more small-molecule drugs targeting PKA and/or CREB, and evaluate their safety and efficacy in cancer therapy. In addition, repurposing some drugs that can act on PKA and CREB may be valuable to cancer therapy.

Concluding remarks

cAMP signaling pathway is evolutionarily conserved and plays key roles in major physiological and pathological processes. The functions of cAMP signaling in tumors depend on cell type and specific environment. In many types of human tumors, cAMP–PKA–CREB pathway plays a tumor-promoting role. The abnormal activation of this pathway and its interaction with other signaling pathways are closely related to tumorigenesis, invasion, metastasis and drug resistance. In contrast, cAMP and CREB also have tumor-suppressive roles in some types of tumors, such as medulloblastoma and non-Hodgkin’s lymphoma [148–150]. While cAMP may inhibit tumor progression through downregulating cyclin A, Erk or other targets, PKA also reportedly inhibits some tumor-promoting signals. For example, PKA may inhibit the activity of mTORC1, an important tumor-promoter, through phosphorylating the mTORC1 component Raptor on Ser⁷⁹¹ residue [4, 151]. It warrants further studies to determine whether targeting cAMP–PKA–CREB pathway is a feasible strategy to treat cancer and reverse cancer drug resistance.

Acknowledgements

Abbreviations

Authors' contributions

HH, HZ and YJ conceived the review and wrote the manuscript. HZ and QK prepared the figures. JW edited the references.

Funding

This work was supported by Grant 81872388 from the National Natural Science Foundation of China and Grant 2018SCUH0009 from the Fundamental Research Fund for the Central Universities.

Availability of data and materials

Not applicable.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Footnotes

Contributor Information

Hongying Zhang, Email: moc.qq@029828596.

Qingbin Kong, Email: moc.361@tudcnibgniqgnok.

Jiao Wang, Email: moc.qq@xoboaijgnaw.

Yangfu Jiang, Email: nc.ude.ucs@ufgnayj.

Hui Hua, Email: moc.361@ixauhiuhauh.

References