Resolution of cAMP signals in three-dimensional microdomains using novel, real-time sensors.

A large number of hormones, neurotransmitters, and odorants alter cellular behavior by triggering changes in intracellular levels of cAMP. Although the effector proteins that bind cAMP have been identified, it is not known how this one messenger can differentially regulate the activities of hundreds of cellular proteins. The spatial and temporal nature of cAMP signals and, thus, their information content remain largely unknown. We present here a high-resolution method for measuring cAMP signals near the plasma membrane in single cells. Cyclic nucleotide-gated (CNG) ion channels from olfactory receptor neurons have been genetically modified to improve their cAMP-sensing properties. We show how these channels can be used in electrophysiological experiments to accurately measure changes in cAMP concentration near the membrane, where most adenylyl cyclases reside. We have found in several cell types (both excitable and nonexcitable) that cAMP is produced in subcellular compartments near the plasma membrane, and that diffusion of cAMP from these compartments to the bulk cytosol is severely hindered. We also show that a uniform extracellular stimulus can initiate very distinct cAMP signals within different compartments of a simple, nonexcitable cell. Analysis of compartmental models indicates that diffusional restrictions between microdomains (near the membrane) and the cytosol, as well as differential regulation of phosphodiesterase activity, are necessary to explain such distinct signals. Using modified CNG channels as sensors has much greater spatial and temporal resolution than other methods for measuring cAMP, and should help to unravel the complexities of signaling by this ubiquitous messenger. Cyclic AMP (cAMP), the prototypical second messenger, regulates a wide variety of cellular processes. Changes in cAMP concentration transmit information to downstream effectors including protein kinase A (PKA), cyclic nucleotide-gated (CNG) channels, hyperpolarization activated (Ih) channels, and Epac. However, it is largely unclear how differential regulation of cellular targets occurs. The concept of compartmentation emerged over 20 years ago in studies of cardiac myocytes, to help explain how a variety of extracellular stimuli that primarily act through cAMP can have very different downstream effects on the cell. The basis for compartmentation, and indeed, the nature of cAMP signals themselves, have remained mysteries. To understand how these signals function within the cell it is important to answer the following questions: (i) How are cAMP signals localized? (ii) What are the kinetics of cAMP signals in localized domains? and (iii) What information is contained in the amplitude and frequency of cAMP signals? We describe here a high-resolution method for measuring cAMP signals near the plasma membrane, using modified cyclic nucleotide-gated ion channels. This approach was inspired by the field of retinal phototransduction, the best-studied second messenger signaling system, in which elegant biochemical studies have been complemented by real-time measurements of cGMP signals using endogenous cyclic nucleotide-gated (CNG) channels. CNG channels are directly opened by the binding of cyclic nucleotides. They were discovered in retinal photoreceptor cells and olfactory receptor neurons, where they generate the electrical response to light and odorants. The native retinal channel is cGMP specific, while the native olfactory channel is equally sensitive to cAMP and cGMP. Native CNG channels consist of A and B subunits, both of which bind cyclic nucleotides, although most A subunits form functional channels on their own. We have modified an olfactory channel A subunit (CNGA2) to improve its sensitivity and selectivity for cAMP. Two of the findings with this approach, summarized here, are: (i) cAMP in several cell types is produced in subcellular compartments under the plasma membrane with restricted diffusional access to the bulk cytosol; and (ii) the amplitude and kinetics of cAMP signals within these compartments are distinct from those in the remainder of the cell.