28.8: Receptor Guanylyl Cyclases, cGMP, and Protein Kinase G

Search Fundamentals of Biochemistry

Learning Goals 

(Learning goals written by Claude, Sonnet 4.6, Anthropic)

Nitric Oxide: Synthesis, Diffusion, and Retrograde Signaling

• Describe the domain architecture of nitric oxide synthase (NOS) isoforms—including the N-terminal oxidase domain (heme-binding), the calmodulin-binding linker, and the C-terminal reductase domain (FMN/FAD/NADPH subdomains)—and explain how Ca²⁺/calmodulin binding enables electron flow from NADPH through FAD and FMN to the heme, driving the two-step monooxygenase conversion of arginine to citrulline and NO.
• Explain how NO's small size and nonpolarity allow it to diffuse across cell membranes without a carrier, enabling retrograde signaling from postsynaptic to presynaptic neurons and paracrine signaling from vascular endothelial cells to adjacent smooth muscle cells, and contrast this mechanism with conventional receptor-mediated signal transduction.
• Distinguish among the three NOS isoforms (nNOS, iNOS, eNOS) with respect to tissue localization, molecular weight, domain structure, and Ca²⁺ dependence, noting that iNOS is constitutively active at basal Ca²⁺ concentrations while nNOS and eNOS require Ca²⁺/calmodulin for activation.

cGMP Synthesis, Degradation, and Pharmacological Targeting

• Compare the two sources of intracellular cGMP—soluble guanylyl cyclase (sGC), a heme-containing heterodimer activated by NO through a coiled-coil conformational change, and particulate guanylyl cyclase (pGC), a single-pass integral membrane receptor activated by natriuretic peptides (ANP, BNP)—and explain how each links its respective upstream signal to cGMP production and downstream vasodilation.
• Explain how cGMP levels are terminated by phosphodiesterases (PDEs) that cleave cGMP to GMP, and describe the pharmacological rationale for PDE inhibitors (sildenafil, tadalafil), sGC stimulators (riociguat), and sGC activators in treating pulmonary hypertension, erectile dysfunction, and benign prostatic hyperplasia, distinguishing among the three drug classes by their dependence on heme redox state and NO availability.

Protein Kinase G: Mechanism and Physiological Functions

• Describe the activation mechanism of PKG: explain how cGMP binding to the two regulatory CAP-Ed domains of the homodimeric kinase relieves N-terminal autoinhibition of the catalytic domain, enables autophosphorylation, and allows substrate phosphorylation, drawing explicit parallels to and contrasts with the activation of PKA by cAMP.
• Distinguish between PKG1 (cytoplasmic, regulates Ca²⁺ homeostasis, smooth muscle relaxation, platelet activation, and neural function) and PKG2 (membrane-anchored via N-terminal myristoylation, regulates bone growth, intestinal secretion, and synaptic plasticity), and explain how PKG-mediated phosphorylation of VDCC, PMCA, IP3R, and SERCA collectively lowers cytosolic Ca²⁺ to produce smooth muscle relaxation and vasodilation.

Much of this material is derived from Friebe et al., cGMP: a unique 2nd messenger molecule – recent developments in cGMP research and development. Naunyn-Schmiedeberg's Archives of Pharmacology volume 393, pages 287–302 (2020). Creative Commons Attribution 4.0 International License. http://creativecommons.org/licenses/by/4.0/

Introduction

We have considered many signal transduction pathways, starting with an extracellular signal, a primary messenger, that initiates signaling when it binds to a receptor (GPCR, RTK, Cytokine Receptor, etc). It elicits a conformational change in the receptor, which is transmitted to an intracellular domain, from which it can propagate a signal intracellularly through secondary messengers and phosphorylation of signaling proteins. Some primary messengers, however, actually pass through the cell either passively or through membrane carriers, so generating a second messenger is unnecessary. We will consider two signals that cross the cell membrane: gases such as nitric oxide (NO) and steroid hormones (which we will discuss in a future section). NO produced within a cell activates the formation of an intracellular second messenger, cyclic GMP (analogous to cAMP). cGMP, in turn, activates our final member of the AGC Ser/Thr protein kinase family, protein kinase G (PKG). Intracellular NO has an unexpected role in adjacent cells. Given its small size and nonpolar nature, it can diffuse from the cell where it is produced into an adjacent cell, thereby initiating signaling. Some call this retrograde signaling.

Before considering cGMP and PKG, let's see how nitric oxide (NO) is produced.

NO formation

NO is synthesized by the enzyme nitric oxide synthase (NOS). There are three isoforms in mammals: neuronal (NOS1 or nNOS), inducible (NOS2 or iNOS), and endothelial (NOS3 or eNOS). Each is a homodimer with a complex domain structure, including a

• N-terminal oxidase (NO_synthase) domain that binds heme
• calmodulin binding site between the N- and C-terminal domains
• C-terminal reductase domain containing the FMN (Flavodoxin) subdomain (which contains an autoinhibitory helix) and FAD/NADPH subdomain.

Ca²⁺ ions activate the enzyme through binding to CAM. A more detailed description of the domain structures of human NOS is shown in Figure \(\PageIndex{1}\). The dimeric molecular weight of the neuronal NOS1 (321K) is greater than for iNOS (206) and eNOS (266), as it has an N-terminal PDZ domain.

neuronal NOS 1 (MW 321)	purple NADPH binding
inducible (MW 206K) endothelial (MW 266)	yellow NADPH binding

Figure \(\PageIndex{1}\): Domain structure of human nitric oxide synthases

NOS catalyzes the conversion of the free amino acid arginine to citrulline and NO, as shown in the chemical equation in Figure \(\PageIndex{2}\).

Figure \(\PageIndex{2}\): Synthesis of NO by nitric oxide synthase

The structures of the three enzymes with or without bound CAM are similar. Linkers between the domains and subdomains allow flexibility. Figure \(\PageIndex{3}\) shows the flow of electrons from NADPH into the reductase (NADPH/FAD subdomain to the FMN subdomain), and on to the NOS synthase domain containing the heme.

Figure \(\PageIndex{3}\): Electron transfer in nitric oxide synthase. Gephart et al. Nitric Oxide (2019). https://doi.org/10.1016/j.niox.2019.04.007. Creative Commons license

Two monooxygenase reactions occur in NOS synthase (oxidative domain), in which electrons are funneled into the heme-bound dioxygen (O₂), forming water and the final product, NO. Electron transfer only occurs within the dimer when calmodulin is bound. However, iNOS is active even at basal Ca²⁺ concentrations.

Figure \(\PageIndex{3}\) shows an interactive iCn3D model of the structure of human neuronal nitric oxide synthase (with its PDZ domain) predicted by AlphaFold (P29475).

Figure \(\PageIndex{10}\): Human neuronal nitric oxide synthase predicted by AlphaFold (P29475) (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...RCgzaSGpc259t6

In the iCn3D model, orient the protein as shown in the figure above. The dark blue (left) is the PDZ domain, and cyan is the oxidase (NOS synthase) domain that contains the heme (which is not shown in AlphaFold models). The spacefill CPK color shown in the cyan domain indicates active-site residues interacting with the heme (not shown). The orange domain is the FMN (Flavodoxin_1) subdomain in the reductase domain. The magenta (far right) shows the FAD/NADPH subdomains of the reductase domain. The yellow spacefill shows the NAD⁺ binding pocket, and the white spacefill shows the FAD binding pocket. The structures of amino acids 129-304 between the PDZ and the oxidase domains are not predicted with any certainty. Crystal structures are available for the oxidase domain alone. The spacefill CPK-colored helix (amino acids 730-754) represents a helical peptide region that binds to calmodulin.

NO, which is synthesized in the cells, can signal there or diffuse to another cell and signal there. Figure \(\PageIndex{11}\) shows how NO, synthesized in vascular epithelial cells that line blood vessels, can move into the nearby muscle cells and initiate signaling through soluble guanylyl cyclase there, leading to vasodilation and a lowering of blood pressure.

Figure \(\PageIndex{11}\): . Mechanisms of nitric oxide-mediated vasodilation. Khalaf et al. Nutrients 2019, 11, 1679; doi:10.3390/nu11071679 CC BY 4.0

Endothelial nitric oxide synthase (eNOS) in the vascular endothelium forms NO from plasma arginine. Two substrates, O₂ and NADPH, are required along with the cofactors tetrahydrobiopterin (BH₄), FAD, and flavin mononucleotide (FMN). NO diffuses into smooth muscle cells and activates soluble guanylyl cyclase (sGC), increasing cGMP production. cGMP subsequently activates protein kinase G (PKG), resulting in decreased [Ca²⁺] by these mechanisms:

• inhibition of voltage-dependent calcium channels (VDCC), reducing calcium influx;
• activation of plasma membrane calcium ATPases (PMCA), increasing ATP-dependent calcium efflux;
• inhibition of inositol triphosphate receptors (IP3R), reducing calcium release from the sarcoplasmic reticulum (SR) to the cytoplasm;
• activation of sarcoplasmic calcium ATPases (SERCA), increasing the ATP-dependent sequestration of calcium from the cytoplasm to the SR.

Decreased [Ca²⁺] mediates smooth muscle relaxation via the activation of myosin light chain kinase and the inhibition of myosin light chain phosphatase (not shown in the figure), resulting in vasodilation.

Figure \(\PageIndex{13}\) shows retrograde diffusion of NO from an activated post-synaptic neuron back to the presynaptic neuron that excites it on the release of the neurotransmitter glutamate. The post-synaptic cell nNOS synthesizes NO after it's activated by Ca²⁺ inflow and binding to CAM (not shown). NO with the pre-synaptic neuron binds to guanylyl cyclase to produce the second messenger cGMP, which can directly activate other channels, protein kinase G, or phosphodiesterases (PDEs)

Figure \(\PageIndex{13}\): NO Signaling at a Neuronal Synapse. https://www.caymanchem.com/news/nitr...ion-in-the-cns.

Synaptic glutamate release activates postsynaptic NMDA and AMPA receptors (NMDAR, AMPAR), leading to Ca²⁺-induced nNOS activation. NO diffuses back to the presynaptic cell and activates sGC, producing cGMP.  cGMP has many signaling roles, including modulating presynaptic neurotransmitter release. cGMP directly targets several ion channels in the postsynaptic cell. As we saw in the previous chapter, many ion channels are voltage-gated. However, ions (e.g., Ca²⁺, Na⁺) and cyclic nucleotides such as cAMP and cGMP can directly regulate ion channels. The latter channels are called cyclic nucleotide-gated (CNG) channels. NO in the post-synaptic cell also associates with CAPON, an nNOS binding protein, leading to a downstream MAP kinase cascade.

Carbon Monoxide

Everyone knows that high doses of carbon monoxide (CO) are lethal, as it binds to heme Fe²⁺ in hemoglobin and myoglobin with a higher affinity than O₂. Hence, it may surprise you that endogenous CO is a signaling molecule, which now, in retrospect, might make sense given its similarity in chemical structure to NO.

CO is produced through heme oxygenase (HOs). CO can act as a signaling molecule in the neural, cardiovascular, respiratory, gastrointestinal, immune, and reproductive systems. In contrast to the lethal effects of inhaling exogenous CO from incomplete combustion, endogenous CO has anti-inflammatory and antioxidant effects. It can also dilate the vascular system. Other gases, like H₂S, are also signaling agents.  H₂S increases in hibernating animals.  It inhibits mitochondrial function and produces torpor.  It likely does this by reducing cytochrome C, the mobile electron carrier for Complex IV.

cGMP formation

The second (or third) messenger cyclic guanosine monophosphate (cGMP) is synthesized after activation of the enzyme guanylyl cyclase (GC) by nitric oxide. cGMP has many signaling effects in cells, some of which were outlined above. The cytoplasmic soluble GC (sGC) is activated by NO. The membrane-associated "particulate" GC (pGC) form is activated upon binding of natriuretic peptides (NPs) to natriuretic peptide receptors, which are NP-activated integral membrane guanylyl cyclases. The peptide hormones (ANP secreted by the atria and BNP secreted by the ventricles) decrease blood pressure. The membrane form does not require NO for activation. Figure \(\PageIndex{14}\) shows the conversion of GTP to cGMP.

Figure \(\PageIndex{14}\): Conversion of GTP to cGMP

Synthesis of cGMP from soluble GC is activated by NO or molecules like nitrates that can be metabolized to NO. These molecules are called NO donors. Since NO causes vasodilation, they are used to treat angina and hypertension. A class of drugs called stimulators (for example, riociguat) increases cGMP production from sGC in the absence and synergistically in the presence of NO. They are also used to treat hypertension. Another class of drugs, called activators, can activate sGC even when heme is oxidized or absent.  Activators can activate sGC even if heme is oxidized or missing without upstream NO signaling. They are effective even if the heme is oxidized or lost from the NOS catalytic domain.

We saw that cAMP is cleaved to AMP by phosphodiesterase. Likewise, phosphodiesterase (PDE) cleaves cGMP to GMP, attenuating cGMP signaling. Selective drugs targeting PDE are available. These include sildenafil for the treatment of pulmonary hypertension and erectile dysfunction, and tadalafil for benign prostatic hyperplasia (BPH).

Pathways for activation of guanylyl cyclase activity (sGC and pGC) are shown in Figure \(\PageIndex{15}\).

Figure \(\PageIndex{15}\): Pathways for activation of guanylyl cyclase activity (sGC and pGC). NP, atrial natriuretic peptide; BNP, brain natriuretic peptide; cGK, cGMP-dependent protein kinase; cGMP, cyclic guanosine monophosphate; CNG, cyclic nucleotide-gated ion channels; CNP, C-type natriuretic peptide; GTP, guanosine triphosphate; GMP, guanosine monophosphate; NO, nitric oxide; NOS, nitric oxide synthase; NP, natriuretic peptide; PDE, phosphodiesterase; pGC, particulate guanylyl cyclase; PKG, protein kinase G; sGC, soluble guanylyl cyclase. Naunyn Schmiedebergs Arch Pharmacol. 2020; 393(2): 287–302. doi: 10.1007/s00210-019-01779-z. http://creativecommons.org/licenses/by/4.0/.

Soluble guanylyl cyclase (cGC) structure and function

The soluble form of GC is a heterodimer of α and β subunits. The domain structure of guanylyl cyclase is shown in Figure \(\PageIndex{16}\).

Figure \(\PageIndex{16}\): Domain structure of guanylyl cyclase. Stuehr et al. JBC REVIEWS| VOLUME 296, 100336, JANUARY 2021. https://doi.org/10.1016/j.jbc.2021.100336. http://creativecommons.org/licenses/by/4.0/)

It appears that when NO binds to the heme group, a twisted coiled-coil domain extends, activating the catalytic domain. Simulators likely induce similar conformational changes by binding to the top part of the CC domain. Figure \(\PageIndex{17}\) shows an interactive iCn3D model of the human soluble guanylate cyclase in the riociguat (stimulator) and NO-bound state (7D9R)

Figure \(\PageIndex{17}\): Human soluble guanylate cyclase in the riociguat and NO-bound state (7D9R) (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...vm3dLK1o1odZD6

The A chain is dark gray, and the B chain is light gray. The guanylate cyclase domain contributed by each monomer is shown in cyan. A phosphonate GTP analog (labeled G2P) is shown in spacefill CPK color bound in the guanylyl cyclase domain (c). The heme (HEM) and the stimulator riociguat (GZO) are shown in the HNOBA (Heme NO Binding Associated) domain and shown in spacefill CPK colors.

The conformation change of the A chain of the bent inactive form of guanylyl cyclase (6JT1) to the active extended form (6JT2) is shown in Figure \(\PageIndex{18}\).

Figure \(\PageIndex{18}\): The conformational change of the A chain of the bent inactive form of guanylyl cyclase (6JT1) to the active extended form (6JT2)

The guanylyl cyclase domain is at the top of the figure. The conformational change is somewhat reminiscent of that in apo-calmodulin upon binding of Ca²⁺ ions. However, the central helix is fully intact in guanylyl cyclase's inactive and active forms.

Figure \(\PageIndex{19}\) shows how oxidation of the heme iron and S-nitrosation of the protein that occurs in the presence of reactive oxygen species (ROS) and reactive nitrogen species (RNS) leads to heme loss and inactivation of sGC.

 

Figure \(\PageIndex{19}\): sGC inactivation pathways and potential structural consequences. sGC inactivation pathways and potential structural consequences. When functional sGC heterodimer (top) is exposed to reactive oxygen and/or reactive nitrogen species (ROS and RNS) it can become unresponsive toward NO by undergoing oxidation of its ferrous heme to ferric (red and green parallelograms) and protein modifications like Cys S-nitrosation (SNO). These events alone or in combination may lead to the breakup of the sGC heterodimer, heme loss, and/or rebinding of Hsp90 to the freed sGCβ1 subunit. NO, nitric oxide; PAS, Per-Arnt-Sim; sGC, soluble guanylate cyclase. Stuehr et al. ibid.

pGC structure and function

The other source of cytosolic cGMP is particulate guanylyl cyclase (pGC). These are integral membrane protein receptors for natriuretic peptides (NPs), which activate the receptor's cytoplasmic guanylyl cyclase domain upon binding. In effect, they are ligand-gated (NP)-receptor enzymes. The peptide hormones (ANP secreted by the atria and BNP secreted by the ventricles) decrease blood pressure. There are seven variants of pGC (A-G) found in mammals. GC-A (also called NPR-A or NPR1) and GC-B (NPR-B or NPR2) are both receptors for natriuretic peptides. The domain structure of NPR-A is shown in Figure \(\PageIndex{20}\).

Figure \(\PageIndex{20}\): Domain structure of GC-A (also called NPR-A or NPR1)

Figure \(\PageIndex{21}\) shows an interactive iCn3D model of the human atrial natriuretic peptide receptor1 AlphaFold predicted model (P16066)

Figure \(\PageIndex{21}\): Human atrial natriuretic peptide receptor1 AlphaFold predicted model (P16066) (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...Wd3xvHrY3rEUw6

Orient the model as shown in the figure above. The coloring in the model is as follows:

• green: Ligand (ANP)-binding domain of the type A natriuretic peptide receptor (NPR-A);
• magenta: PK-GC pseudokinase domain;
• cyan: cyclase domain;
• rainbow helix: HNOBA domain is found to be associated with the HNOB domain and pfam00211 in soluble cyclases and signaling proteins. The HNOB domain is predicted to function as a heme-dependent sensor for gaseous ligands;
• gray helix: transmembrane segment amino acids 474-494.

Note that the protein is an integral membrane protein that passes through the membrane using a single alpha helix (474-494). The N-terminal domains above it and the C-terminal domains below it would orient like a typical single-pass membrane protein in the presence of a bilayer.

Protein Kinase G (PKG)

To briefly review, NO production leads to the production of cGMP. cGMP can directly bind to and regulate membrane ion channels. In line with the basic signaling paradigm described throughout this chapter, we will now discuss how it activates Protein Kinase G (PKG), a member of the AGC family of Ser/Thr protein kinases.

There are two mammalian genes for PKG1 and PKG2. Both are homodimers. PKG1 acts in the cytoplasm while PKG2 becomes tethered to the membrane by N-terminal myristoylation. Figure \(\PageIndex{22}\) shows the domain structure of PKG-I, which is similar to PKG2.

Figure \(\PageIndex{22}\): Domain structure of protein kinase G-I.

Red indicates the two nonidentical cGMP-binding domains. Green is the N-terminal coiled-coil dimerization domain, which inhibits kinase activity in the absence of cGMP. On binding of cGMP, autoinhibition of the catalytic domain by the N-terminal domain is relieved. The binding of cGMP to the regulatory domain induces a conformational change that relieves the N-terminus's inhibition of the catalytic core and allows the phosphorylation of the regulatory domain (autophosphorylation) and of substrate proteins. Whereas PKG-I is predominantly localized in the cytoplasm, PKG-II is anchored to the plasma membrane by N-terminal myristoylation.

PKG1 is involved in modulating Ca²⁺ activity, platelet activation, smooth muscle contraction, gene expression, and neural function. PKG2 helps regulate bone growth, intestinal secretion, and synaptic plasticity. It also regulates gene expression and activates the MAPK cascade in bone cells.

Figure \(\PageIndex{23}\) shows an interactive iCn3D model of the predicted structure of Human cGMP-dependent protein kinase 2 (AlphaFold, Q13237).

Figure \(\PageIndex{23}\): Predict the structure of Human cGMP-dependent protein kinase 2 (AlphaFold, Q13237). (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...A55VorQFGZqpSA

The cyan domain is the STK (Ser/Thr Kinase) domain. The magenta and purple domains are the CAP-Ed domains. The spacefill side chain structures within them represent the cGMP (or with lower affinity cAMP) binding sites. The CPK-colored sticks in the cyan kinase domain are the amino acid side chains in the active site where ATP binds. The orange backbone represents the least confident part of the predicted structure. The black spacefill is the N-terminal Gly (after removal of Met), which is myristoylated, allowing targeting of the modified PKG2 to the cell membrane.

Summary

(Summary written by Claude, Sonnet 4.6, Anthropic)

Nitric oxide (NO) represents a distinct signaling paradigm in which the primary messenger is itself a small, nonpolar gas that bypasses membrane receptors entirely, diffusing freely across lipid bilayers to act within the cell where it is produced or in adjacent cells. NO is synthesized by nitric oxide synthase (NOS), a homodimeric enzyme that channels electrons from NADPH through FAD and FMN cofactors in its C-terminal reductase domain to the heme-containing N-terminal oxidase domain, where O₂ and arginine are converted to citrulline and NO in two sequential monooxygenase reactions. The three mammalian NOS isoforms—nNOS (neuronal), iNOS (inducible), and eNOS (endothelial)—differ in tissue distribution, molecular weight, and Ca²⁺ dependence: nNOS and eNOS require Ca²⁺/calmodulin for electron transfer between subunits, while iNOS is constitutively active at basal Ca²⁺ levels. Endogenous CO and H₂S are additional gaseous signaling molecules, illustrating that gas-phase signal transduction is a broader biological strategy.

NO diffuses into target cells and binds to the heme of soluble guanylyl cyclase (sGC), a heterodimeric cytoplasmic enzyme whose activation involves a coiled-coil conformational change that activates the catalytic domain to convert GTP to the second messenger cGMP. A pharmacologically important alternative source of cGMP is particulate guanylyl cyclase (pGC), a single-pass integral membrane receptor for natriuretic peptides (ANP and BNP) that activates an intracellular guanylyl cyclase domain on peptide binding. cGMP is terminated by phosphodiesterases (PDEs), and selective PDE inhibitors—including sildenafil and tadalafil—exploit this pathway therapeutically. sGC stimulators, such as riociguat, increase cGMP production in the presence or absence of NO, while sGC activators bypass the requirement for intact heme, providing therapeutic options when oxidative stress has inactivated the enzyme.

cGMP activates the final member of the AGC Ser/Thr kinase family, Protein Kinase G (PKG). Like PKA, PKG exists in an autoinhibited state in the absence of its cyclic nucleotide activator; cGMP binding to two regulatory CAP-Ed domains relieves N-terminal inhibition of the catalytic domain and enables autophosphorylation and substrate phosphorylation. PKG1 operates in the cytoplasm and mediates smooth muscle relaxation by phosphorylating multiple Ca²⁺-handling proteins—inhibiting VDCC and IP3R to reduce Ca²⁺ entry and ER release, while activating PMCA and SERCA to increase Ca²⁺ efflux and sequestration—thereby lowering cytosolic [Ca²⁺] and promoting vasodilation. PKG2 is anchored to the plasma membrane via N-terminal myristoylation and regulates bone growth, intestinal secretion, and synaptic plasticity. Together, the NO→sGC→cGMP→PKG axis illustrates how a gaseous paracrine signal can initiate a precisely controlled intracellular kinase cascade with broad physiological and therapeutic significance.