Putting on the Brakes: Cyclic AMP as a Multipronged Controller of Macrophage Function

See allHide authors and affiliations

Science Signaling  16 Jun 2009:
Vol. 2, Issue 75, pp. pe37
DOI: 10.1126/scisignal.275pe37


Macrophages orchestrate innate immune responses in tissues by activating various proinflammatory signaling programs. A key mechanism for preventing inflammatory disease states that result from excessive activation of such programs is the generation of the second messenger cyclic adenosine monophosphate (cAMP) by ligation of certain guanine nucleotide–binding protein (G protein)–coupled receptors (GPCRs). The pleiotropic actions of this cyclic nucleotide on various inflammatory functions of macrophages are mediated by diverse molecular mechanisms, including the assembly of distinct multiprotein complexes. A better understanding of crosstalk between cAMP signaling and proinflammatory pathways in macrophages may provide a basis for improved immunomodulatory strategies.


Macrophages are the resident immune cells of tissues and provide the first line of defense against infection. Cyclic adenosine monophosphate (cAMP) is the founding member of the family of second messenger molecules (1) and is a universal regulator of cellular function (2). The regulation of macrophage function by cAMP has important pathophysiologic, clinical, and therapeutic relevance. One of the compelling challenges in cell signaling research is to address the question of how a single cAMP signaling pathway influences such a diverse array of cellular functions. Studies into the mechanisms by which this cyclic nucleotide acts on macrophages, described in an accompanying article by Wall and co-workers (3), emphasize that the answer to this question is, quite simply, that there is not one cAMP pathway, but many.

Macrophages in Innate Immune Responses

Upon their first encounter with pathogens such as bacteria or fungi, macrophages endeavor to phagocytose (living up to their moniker of “big eater”) and then kill the interlopers. Evolution has endowed the macrophage with the additional capabilities to recognize a diverse spectrum of microbial surface molecules through pathogen-recognition receptors [for example, Toll-like receptors (TLRs)], to generate substances, such as complement, chemokines, and leukotriene B4, which are capable of recruiting phagocytic reinforcements (such as neutrophils) from the bloodstream, and to activate cellular programs that amplify their own and their neighboring cells’ antimicrobial arsenals (4). A prominent component of these programs is the family of genes, including those that encode tumor necrosis factor-α (TNF-α), macrophage inflammatory protein-1α (MIP-1α), interleukin-8 (IL-8), and inducible nitric oxide synthase (iNOS), whose expression is controlled by the transcription factor nuclear factor-κB (NF-κB) (5). This extensive functional repertoire enables the macrophage to coordinate a physiologic inflammatory response designed to defend the host (Fig. 1). If, however, this inflammatory response exceeds that which is required to eliminate the infection, persists after the pathogen is cleared, or is inadvertently activated by nonmicrobial substances that are mistakenly perceived as microbes, pathologic inflammatory states ensue. It is therefore critical that the proinflammatory potential of the innate immune system and of macrophages in particular be tightly controlled, and cAMP can be considered the master controller of the innate immune system.

Fig. 1

Regulation of innate immune activation pathways in macrophages by cAMP signaling. Microbial components or opsonins engage macrophage receptors that mediate phagocytosis and killing of the pathogen, as well as other receptors that promote the generation of cytokines through, for example, activation of NF-κB. Bacterial toxins (through direct actions on G proteins) or host-derived substances produced during infection, such as PGE2 and adenosine (through ligation of their GPCRs), enhance the generation of cAMP. cAMP modulates all of the innate immune functions of macrophages by activating PKA, Epac, or both. The inhibition of TNF-α production by cAMP is mediated by PKA II anchored to AKAP95, which prevents the activation of NF-κB. Inhibition of the production of MIP-1α by cAMP is likewise mediated by PKA II, which is anchored to an as-yet-unidentified AKAP.

The Cast of Characters That Constitutes the cAMP System

Ligand binding to a guanine nucleotide–binding protein (G protein)–coupled receptor (GPCR) that is coupled to a stimulatory G protein α-subunit (Gαs) activates adenylyl cyclase (AC) to catalyze the formation of cAMP from adenosine triphosphate (6). Examples of such ligands include epinephrine, adenosine, vasoactive intestinal polypeptide, and certain prostaglandins. The concentration of intracellular cAMP is also regulated by the enzyme phosphodiesterase (PDE), which degrades cAMP. The transmission of cellular information by cAMP has usually been attributed to the classical effector cAMP-dependent protein kinase (PKA). Upon the binding of cAMP to the regulatory subunits of PKA, the catalytic subunits of this serine/threonine kinase are released and phosphorylate various targets, including signal transduction proteins as well as the transcription factor cAMP response element–binding (CREB) protein (7). Although the canonical paradigm of cAMP→PKA→CREB has become ingrained as signaling dogma, many important observations over the past decade instead suggest a far more complex signaling network.

First, an alternative effector, guanine nucleotide exchange protein activated by cAMP (Epac)—which activates the small guanosine triphosphatase (GTPase) Rap1—is implicated in certain actions of cAMP in various cell types (8), including macrophages. Interestingly, PKA and Epac may have redundant, independent, or even opposing effects within the same cell. Second, there are numerous molecular species of many of the pathway components, including two Gαs proteins, 10 isoforms of AC, more than 50 PDEs, two types of PKA, and two isoforms of Epac. Of course, many different GPCRs funnel into this pathway, and some ligands [for example, prostaglandin E2 (PGE2) and adenosine] are recognized by more than one Gαs-coupled GPCR. Even though particular component species are expressed in a cell-specific fashion, it is likely that hundreds of combinatorial pathways exist in any given cell type. Third, different species of pathway proteins within a given cell type may be localized to distinct subcellular sites (9). One mechanism of localization involves posttranslational modifications of pathway components that permit interactions with certain membranes or membrane microdomains. Another mechanism—which is central to the findings of Wall and colleagues—involves the assembly of pathway components into multiprotein signaling complexes that are organized by scaffold proteins termed A-kinase anchoring proteins (AKAPs) (10, 11). AKAPs are a group of more than 50 structurally diverse proteins that share the ability to anchor PKA to specific membrane constituents, thereby focusing its kinase activity on discrete pools of substrate. AKAPs also bind to PDEs, which contributes to spatially and temporally localized gradients of cAMP within a cell. Finally, because they also bind to other signaling enzymes, including both signal transduction proteins [kinases, small GTPases, and even Epac (12)] and signal termination proteins (phosphatases and GTPase-activating proteins), these complexes have the capacity to integrate cAMP signaling with other signaling cascades. This is critical to appreciating the work of Wall and colleagues (3).

cAMP and Macrophage Function

PGE2 is likely to be the most important of the ligands of Gαs-coupled GPCRs in the context of innate immunity because it is produced in abundance at sites of infection and it inhibits the key proinflammatory functions of macrophages through cAMP-dependent mechanisms. Such actions of PGE2 and cAMP include (i) inhibiting phagocytosis mediated by multiple recognition receptors; (ii) inhibiting microbial killing by interfering with the generation of reactive oxygen species (ROS) and antimicrobial peptides; and (iii) inhibiting the generation of proinflammatory mediators such as TNF-α, MIP-1α, and leukotriene B4 while enhancing the generation of the antiinflammatory cytokine IL-10 and the suppressor of cytokine signaling-3. Limited data are available regarding the relative roles of PKA and Epac in suppressing these various functions of macrophages. The most comprehensive data set in a single population of macrophages involves the primary pulmonary alveolar macrophage. In this cell type, PKA modulates the generation of mediators (inhibition of TNF-α, MIP-1α, and leukotriene B4 and enhancement of IL-10 and IL-6), Epac mediates the inhibition of phagocytosis of antibody-coated targets, and both PKA and Epac mediate the inhibition of bacterial killing by the generation of ROS (13) (Fig. 1). However, these same authors found that, by contrast, generation of dendritic cell mediators is modulated by both PKA and Epac (14), attesting to the importance of cell specificity in the roles of distinct cAMP effectors.

cAMP- and PKA-Mediated Regulation of TLR4 Signaling and the NF-κB Pathway

The best understood pathogen recognition receptor is TLR4, which recognizes the lipopolysaccharide (LPS) of gram-negative bacteria and transduces signals through various kinases, including mitogen-activated protein kinases, and NF-κB to mediate profound inflammatory responses, including septic shock (15). Anti-inflammatory and immunosuppressive actions of cAMP have been recognized for 35 years (16), and these effects reflect the inhibition of various kinases and of the activation of NF-κB. Wall and co-workers (3) have carefully dissected the mechanisms by which cAMP modulates LPS–induced expression of the genes encoding TNF-α, MIP-1α, granulocyte colony–stimulating factor (G-CSF), and IL-10 in macrophages. Their studies employing PKA- or Epac-selective cAMP analogs showed that each of these four cytokines was regulated by PKA but not by Epac. Their subsequent findings emphasize the diversity of operative mechanisms by which PKA controls different cytokines, break new ground regarding the roles of AKAP in such processes, and also extend our understanding of how PKA can regulate the activation of NF-κB.

Wall et al. identified three different mechanisms by which PGE2 or cAMP modulated LPS-induced cytokine production. First, enhanced generation of IL-10 was independent of AKAPs but was unique among the four cytokines in its dependence on CREB. Second, enhanced generation of G-CSF was dependent on an interaction between type I PKA and an AKAP that was not identified. Third, the suppressed production of TNF-α and MIP-1α depended on an interaction between type II PKA and an AKAP, which in the case of TNF-α was ultimately identified as AKAP95 (Fig. 1).

The authors proceeded to demonstrate that AKAP95 formed a complex with, and directed the PKA II–mediated phosphorylation of, p105 (also known as Nfkb1), a protein that has the potential to function as a cytosolic inhibitory anchor (IκB) of the nuclear translocation of NF-κB. Such phosphorylation interfered with phosphorylation by IκΒ kinase and the subsequent degradation of p105, events required for LPS-induced activation of NF-κB. Although it remains to be determined whether this same mechanism accounts for the inhibition of MIP-1α by PGE2 and cAMP, this new work provides the first link between AKAPs and the control of NF-κB signaling.

These new observations exemplify the insights to be gained by further dissection of the crosstalk between the cAMP axis and TLR4 signaling, as well as NF-κB activation. However, as the actions of PGE2 and cAMP in controlling innate immune responses are not confined to a single recognition receptor, proinflammatory signaling pathway, or cellular function, there are many more mechanistic questions to be addressed. How are the other multicomponent signaling complexes involved in the many actions of cAMP organized? How is information flow directed preferentially to Epac rather than to PKA? Do different Gαs-coupled GPCRs couple to functionally different signal transduction pathways? To what extent is the cAMP signaling machinery used in transducing the actions of Gαi-coupled GPCRs that inhibit AC? The pathophysiologic circumstances that result in perturbation of the concentration of intracellular cAMP are myriad, reflecting the actions not only of host-derived molecules such as PGE2 but also of pathogen-derived molecules (for example, cholera and pertussis toxins) and pharmacologic agents (for example, aspirin-like drugs and β-adrenergic agents used as bronchodilators in asthma). A better understanding of how the cAMP axis controls innate immunity therefore also has implications for the development of immunomodulatory therapeutics.


View Abstract

Stay Connected to Science Signaling

Navigate This Article