Adenylyl cyclase

Adenylyl cyclase (ADCY, EC number 4.6.1.1), also known as adenylate cyclase, is an enzyme which catalyzes the cyclization of into ^[1]. The occurs in a single, concerted step, where the oxygen on ATP's 3' hydroxyl group nucleophillically attacks the alpha-phosphate forming a phosphodiester bond and cleaving a pyrophosphate group^[2]. In most active sites there is an acidic residue near the 3'OH which functions in its deprotonation, and basic residues by the β-phosphorous to lower the energy of the group for cleavage^[3][4]. The reactant in the reaction catalyzed by adenylyl cyclase is ATP; ATP is the most abundant nucleotide triphosphate in most cells with typical concentrations ranging from 1 to 10mM^[5]. This high intracellular concentration allows for cAMP concentrations to rise quickly in response to a specific signal, which is important in many signal transduction and metabolic pathways^[6]. The main product of this reaction is cAMP, with a side product of PP_i^[1]. The cytoplasmic regions of ADCY consist of its N-terminal, C1a, C1b, C2a and C2b. The C1a and C2a make the catalytic domain. This page emphasizes on the microbial adenylyl cyclase Rv1264, but mammalian adenylyl cyclases are also covered in less detail.

Introduction

There are ten isozymes of adenylyl cyclases in mammals, adenylyl cyclase type I-X, (ADCY I-X), and many more in other organisms^[1]. All mammalian, and most other adenylyl cyclases belong to class III; most are integral membrane proteins, and all produce cAMP, the ability of which can be activated or inactivated in response to certain conditions or ligands^[1]. All mammalian membrane bound adenylyl cyclases are activated by alpha subunits of G-proteins, but respond differently to ligands such as magnesium ions, calcium ions, and beta gamma subunits of G proteins^[1]. One of the mammalian isozymes, and some prokaryotic forms of adenylyl cyclase respond to environmental conditions, primarily pH^[7][8].

Cyclic Adenosine Monophosphate

In mammals, cAMP acts as a secondary messenger, one of its functions is to control the activity of protein kinase A (PKA)^[9]. In turn, PKA has quite diverse roles in cells, while most of them are associated with metabolism, PKA also plays important roles in transcription, the cell cycle, and apoptosis^[9]. The ultimate fate of cAMP is its transformation into AMP by the cleavage of the phosphodiester bond by 3', 5'-cyclic adenosine monophosphate phosphodiesterase^[9].

Pyrophosphate

Cleavage of the by-product of this reaction, PP_i, by pyrophosphatase yields two molecules of inorganic phosphate (P_i)^[10]. ATP synthase can reincorporate this inorganic phosphate into adenine diphosphate (ADP) to make ATP using energy in the proton motive force^[11].

Mammalian Adenylyl Cyclase

There are ten isozymes of adenylyl cyclases in mammals, adenylyl cyclase type I-X, (ADCY I-X); In mammals adenylyl cyclase plays an important role in signal transduction pathways in which cAMP is a secondary messenger^[12].

ADCY I-IX all share a general structure; They are composed of two trans-membrane regions (M1, M2) which are composed of six membrane-spanning helices and function to keep the enzyme anchored in the membrane, and two cytoplasmic regions (C₁, C₂) which can be further sub divided (C_1a, C_1b, C_2a, C_2b) and are responsible for all catalytic activity, and regulation by G-proteins and forskolin^[12]. In solution, the C_1a and C_2a domains can form heterodimers with each other, either in the same or different enzymes, or they can form homodimers with their identical units on different enzymes^[3]. The C_1b domain is very large (≈15 kDa) with many regulatory sites, and has a variable structure across isozymes; while the C_2b domain is nearly non-existent in many isozymes, and has yet to be associated with a particular function^[3].

Type II

Structure

A monomer of C₂ domain of has an internal, hydrophobic, anti-parallel surrounded by several, amphipathic , except for an area which needed to form a homodimer with another C₂ domain^[3]. Two monomers of C₂ domains of type II adenylyl cyclase bind together in solution to form a , which is necessary for catalytic conversion of ATP to cAMP and PPi^[2]. When they are bound they create a deep crevasse spanning the center of their binding site; this crevasse is suited to bind two molecules at its ends^[13]. Strong hydrogen bonds are made between oxygen atoms of forskolin and the surrounding peptide backbone, and the rest of the interactions are highly hydrophobic, as the forskolin binding site contains ten aliphatic and aromatic residues^[13]. This binding of forskolin creates a hydrophobic linkage between the monomers, each of which has two different hydrophobic surfaces binding to forskolin; and it is this interaction which makes the homodimer stable^[3][13]. The forskolin also interacts with and properly positions Asn 1025, which is essential for catalytic activity, and it may even interact directly with the ATP^[13]. This homodimer-forskolin complex can be further activated in response to a signal via binding to a G-protein’s βγ-subunit ^[2][13]. This βγ-subunit binds to which make up part of an α-helix on the outermost layer of the complex^[2][14].

The of this homodimer is located within the crevasse, and is characterized by two highly conserved sets of polar residues (Arg 997 (Green), Asn 1025 (Red), Ser1028(Pink), Arg 1029(Orange), Asp 1031(Yellow), and Ser 1032(Purple))^[2][3]. One of these sets is located on each monomeric subunit, on the homodimer they arrange themselves in an anti-parallel fashion, where they point towards each other^[3].

Over-expression Disorders

In the brain of mammals, the execution of memory based functions is carried out by the prefrontal cortex (PFC)^[15]. Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels on neurons close to allow electrochemical signals to flow down the axon and into a synapse; when HCN channels are open, the electropotential signal cannot be transmitted through the cell^[15]. Exposure of these channels to cAMP causes them open, stopping the transmission of signals, and thus impairs higher cognitive thoughts^[15]. In patients with schizophrenia, a cAMP regulatory molecule, Disrupted-in-Schizophrenia 1 (DISC1) is mutated and cannot regulate cAMP levels^[15]; Thus, elevated cAMP levels may cause schizophrenia. The closing of HCN channels are thought to play a part in other disorders, such as ADHD and bipolar disorder^[15]. Thus, it is reasonable that regulation of cAMP production by targeting type II adenylyl cyclase, since it is found in the brain, may act as a treatment for these disorders.

Rv1264 Adenylyl Cyclase

Although adenylyl cyclase is found throughout organisms at a universal level, distantly related organisms have different modifications of the enzyme, each is specialized for a particular task in a particular environment^[16]. As stated earlier humans have 10 known isozymes of adenylyl cyclase; whereas Escherichia coli has only one isozyme, and Mycobacterium tuberculosis has 15^[16]. One particularly interesting adenylyl cyclase possessed by M. tuberculosis, Rv1264, has a N-terminal which, in a sense, acts as a pH sensor, as it regulates the activity of the enzyme based on the pH of the surrounding solution^[8][16]. This adenylyl cyclase, like most others, belongs to class III, adenylyl cyclases in this class have multiple domains, at least one for catalysis, and another for regulation^[16].

Structure

The adenylyl cyclase is a 363 residue long protein composed of a catalytic and a regulatory which contains a flexible which connects the two^[16]. The active structure is a homodimer resembling the mammalian type II homodimer, where an α-helix of one monomer (Chain A) is placed through the central coiled coil of another (Chain B)^[3][16]. This dimerization places the regulatory domain of chain A in close proximity to the catalytic domain of chain B, and vice versa^[16]. Two switch elements are present in the protein, one in the C-terminal domain (α1-switch) and another in the linker region (αN10-switch), these allow for large conformational changes to take place in the enzyme in response to relatively small environmental changes^[8][16].

C-Terminal Catalytic Domain

The catalytic activity in Rv1264's is performed by the residues: Asp 222 (Red), Lys 261 (Blue), Asp 265 (Orange), Arg 298 (Pink), Asp 312 (Yellow), Asn 319 (Purple), Arg 323 (Green)^[16]. All of these residues create a highly polar environment which is complementary in charge and polarity to the intermediate of the reaction^[8][16]. Residues which guide the phosphates of ATP, arginine 298 and 323, bind a in the active site^[17]. This sulphate ion is located in the position which will be occupied by the ATP's β-phosphate during catalysis^[17]. In the process of catalysis the β-phosphate is cleaved from the α-phosphate; this reaction may be made more favourable by lowering the energy though complementary charge associations between the β-phosphate and arginines 298 and 323^[16][17]. Another residue, , binds a molecule through electrostatic interactions^[16]. The specific function of this association is unknown, but because of its close proximity to the active site it may play a role in catalysis^[16].

Amino acid sequencing has shown that the sequence between the Rv1264 adenylyl cyclases catalytic domain and mammalian adenylyl cyclases catalytic domain are not well conserved, with only about a 25% correspondence to the mammalian type II adenylyl cyclase discussed above^[8][16]. The two different isozymes; however, still resemble each other in that superimposition of one over the other has a substantial overlap, with a root mean square deviation (rmsd) of less than 1.76Å between 79% of all ^[8][16]. A notable difference in Rv1264 and type II adenylyl cyclase catalytic domains is their relative sizes; Rv1264 has no dimerization arm and several of its loops have been shortened^[8][16]. This results in Rv1264 adenylyl cyclase catalytic domain associating as a dimer with a smaller interface area than mammal's type II, with interface areas of 1900Ų and 3800Ų, respectively^[16]. The active sites the Rv1264 and mammalian type II adenylyl cyclases are even more conserved; the in the active site have an rmsd of only 0.69Å, and in the active site have a rmsd of 1.17Å^[8][16].

All of the above structural information for the C-terminal catalytic domain is only true when the enzyme is in its active state. The inactive state of the enzyme possesses a disassembled active site, hence no catalytic activity. Relative to the fixed N-terminal domains, each of the C-terminal monomeric domains can transpose up to 6Å and rotate by 55^o[16]. This massive change in the C-terminal domains tertiary structure disassembles the active site, moving catalytic residues up to 25Å away from their catalytically active position^[16]. Not only is the active site disrupted in the C-terminal domain, but the interface area between the two monomers decreases in size from 1900Ų to 930Å^2[16].

α1-switch

The is located within the C-terminal catalytic domain and exists as a compact α-helix when the enzyme is in its active state,^[16]. Upon inactivation the α-helical conformation of the α1-switch becomes unstable, and it transforms into a random coil^[16]. Since this switch is in a close proximity to the active site, a major change in structure greatly disrupts the active site, and yields the enzyme inactive^[16].

This structure is conserved in mammalian adenylyl cyclases, where it acts as a contributor for the binding site of the βγ-phosphates of the ATP substrate^[18]. It also functions in regulation of mammalian adenylyl cyclases, along with another α-helix, it forms the binding site for the G_sα and G_iα G protein subunits which regulates the adenylyl cyclases activity^[3].

N-Terminal Regulatory Domain

The N-terminal domain functions in regulation; it has a unique mechanism that determines whether the enzyme will be active or not based on the pH of the surrounding solution^[8]. Each monomer in the catalytically active dimer has ten , when they are dimerized they form a disc like structure^[16]. A single molecule of binds in a hydrophobic pocket^[19]. The function of this polyethylene glycol appears to be structural; however, its specific placement and movement between active and inactive forms of the enzyme suggests that it may also function in detecting hydrophobicty changes in the environment^[19].

αN10-switch

The is located within the linker region and its conformational change has a more drastic effect on the overall enzyme than the α1-switch does^[16][19]. When the enzyme is active, the αN10-switch consists of a random coil with a short α-helix; this conformation allows for only a weak interaction between the N-terminal regulatory domain and the C-terminal catalytic domain, allowing the enzyme to exhibit catalytic activity^[16]. This helix can extend by up to 24Å, which separates the monomeric C-terminal catalytic domains of the dimer^[16]. This separation of domains lowers their interface area and transposes residues; as discussed above, these changes result in an inactivation of the enzyme, and thus, a major characteristic of the inactive state of the enzyme is an extended αN10-switch^[16][19]. As the αN10-switch extends, the moves outwards, and the pentaethylene glycol moves into a newly formed cavity by the αN10 helix^[19]. This further supports the idea that the pentaethylene glycol ligand functions not only for structure, but also for regulation^[19].

Regulation by pH

At the beginning of the linker region there is a residue, , which has a considerable influence on regulation by pH^[16][19]. At a basic pH the residue has no charge and minimal interactions with the two catalytically important residues , which lie 14 Å and 21 Å away from their catalytically active positions^[16][19]. The enzyme is inactive at this state, not only because of the Lys 261 and Asp 312 residues, but also because of other major structural components of the catalytic site being disassembled. At an acidic pH Rv1264 becomes active (Optimal pH~5.8)and has up to a 40 fold increase in activity^[16]. At this acidic pH, His 192 becomes protonated and positively charged; which creates major structural changes throughout the protein, including compaction of both the αN10-switch and α1-switch, contraction of the α4-helix which in turn causes a translocation of the paraethylene glycol ligand^[8][16][19]. The positively charged His 192 electrostatically repels the Lys 261 and Asp 312 residues which along with other structural changes transposes them 14 Å and 21 Å, respectively, into their catalytically active positions^[16].

, a residue which organizes many residues which are important in the C-terminal - N-terminal interaction through hydrogen bonds, is also important for regulation^[16]. When this residue is mutated, regulation based on pH is lost and the enzyme is constantly active^[16].

Many other residues also contribute to regulation by pH; it is the electrostatic interactions and hydrogen bonds in the C-terminal - N-terminal domain interface which allows Rv1264 adenylyl cyclase to be sensitive to pH^[16].