Transmembrane signalling through G proteins

To respond to environmental changes, cells must transfer information concerning their environment across their impermeable membranes. One solution to this problem involves three proteins: a receptor which detects information outside the cell, an effector molecule which alters the intracellular environment, and a guanine nucleotide regulatory protein (G protein) which couples the receptor to the effector molecules. The features of this system that make it particularly useful for transmembrane signalling are that signals can be amplified at several steps, and that the transfer of information is unidirectional. This three protein system is used by organisms ranging from yeast to mammals, and is versatile, responding to signals from light and smell to hormones and neurotransmitters. The signals regulate rapidly occurring events such as action potentials, and long-term events such as growth and development [reviewed in 1–6]. In the kidney, G proteins are involved in the action of hormones such as PTH, vasopressin, and angiotensin II, and the activation of second messenger systems such as adenylyl cyclase, phospholipases, and ion channels. This paper examines the basic mechanisms by which G proteins work, focusing on the α chains, the general signalling pathways in which they are involved, and finally considers how what has been learned about their function can be applied to understanding the kidney.G proteins are heterotrimeric proteins composed of α, β and γ subunits which transfer information from receptors to effectors through cyclic changes in conformation driven by GTP binding and hydrolysis. The various G proteins are characterized by their α subunits because the α subunits are the most diverse, and are generally believed to be responsible for the specificity of receptor and effector interactions. The α subunits are also responsible for GTP binding and hydrolysis, reversible interactions with βγ subunits, and are the substrates for ADP-ribosylation by cholera and pertussis toxins. To date, evidence for sixteen α subunits exists (Table 1), all of which show striking structural similarity to one another [1–8]. When compared to αs, the other α subunits which have been completely cloned (αolf, αo, αi1-3, αt, and αz) are 42 to 46% homologous, but if αs is left out of the comparison, the homology ranges from 60 to 72% [2, 7].The β and γ subunits are less well understood, but are believed to be important for membrane interactions and receptor coupling, and may in some systems be able to activate effectors. to date, four β and up to four subunits have been identified [MI Simon, personal communication]. The β subunits are similar with only 15% divergence at the amino acid level [8, 9, M.I. Simon, personal communication]. Transducin γ and brain Go and Gi-associated γ are 55% homologous [10, 11]. Although βγ subunits purified from retina and brain inhibit cyclase with different efficiency, the functional differences among the β and γ subunits from brain have not been determined [11].G proteins are implicated in a signalling pathway by one or more of several criteria, activation of the pathway by nonhydrolyzable GTP analogs or A1F4-, inhibition by GDPβS, sensitivity to pertussis or cholera toxins, stimulation of GTP hydrolysis by the receptor agonist, and regulation of receptor affinity for agonist by guanine nucleotides [2,3]. Although these criteria serve to distinguish G protein-dependent signalling pathways from others such as tyrosine kinase receptors, or ligand-gated ion channels, they frequently do not help to distinguish among the many possible G proteins.Only two systems have been well characterized at the molecular level. Most of the detailed biochemical information on G protein action comes from the study of the retinal rhodopsintransducin-cGMP phosphodiesterase system, and the Gs-adenylyl cyclase system. In the retinal system, large amounts of the three principal components—rhodopsin, the receptor; transducin, the G protein; and cGMP phosphodiesterase, the effector—can be purified relatively easily in large quantities, allowing detailed biochemical studies [5, 6, 12]. The Gs-adenylyl cyclase system has been accessible to study despite small quantities of protein because a mutant cell line lacking Gs, Cyc-, allowed its identification, purification, and characterization by reconstitution [2, 3].For the other α chains, less precise information is available for a number of reasons. Multiple α chains are present in a single tissue or cell type, they are similar biochemically and difficult to purify to homogeneity, and appropriate mutant cell lines analogous to Cyc- for reconstitution do not exist. Additional problems in identifying the signalling pathways served by a specific G protein are that the components of signalling pathways are present in small quantities in most tissues, and the pathways are complex. Multiple receptors can activate a single G protein, a single G protein can activate multiple effectors, and multiple G proteins may activate the same effectors. Progress has been made with antisera and reconstitution of purified proteins into relatively “complete” systems such as whole cells and membrane patches, but the results are not as clean as for transducin and Gs