Heterotrimeric G proteins 

 The G protein activation/deactivation cycle 

As we have learned previously, activation of a G protein-coupled receptor by its ligand results in conformational changes in the receptor that are transmitted to the G protein. In its inactive state, the G protein exists as and abg trimer with the a subunit bound to GDP. Receptor activation causes the a subunit to lower its affinity for GDP and bind to GTP (which exists at concentrations considerably higher than those for GDP). GTP binding to the a subunit results in a profound change in its conformation and causes the a-GTP to dissociate from the bg dimer and activation of the G protein. Depending on the system, either (or both) the a-GTP or the bg subunits can alter the activity of downstream effector molecules. The intrinsic GTPase of the a subunit then turns off the signal by converting GTP to GDP, whereupon the G protein returns to its resting state (ready to be turned on again through another activated receptor). What I’ve just described is really a cyclic event, where the G protein can turn on and off in a regulated and timed fashion. 

G protein structure & function 

G proteins can be classified on the basis of differences in Ga subunit amino acid sequence, sensitivity to specific bacterial toxins (see below), and type of effector molecule(s) regulated by a particular G protein. Through their extensive diversity, the G protein family provides a good example of how nature can use proteins with similar basic functions to generate a number of different biochemical responses. There are at least twenty different a, five b, and eleven g subunits documented in the literature. Thus, the possibility to form multiple a,b, and g combinations may contribute to a significantly diverse range of responses associated with these different G protein isoforms. It should be noted that the consequences of specific b and g pairings is poorly understood.

·        Three subunits: a, b, and g. 

GTP-binding (G) proteins are major participants in numerous signal transduction pathways. Their activity has been linked to signals that lead to cell division, changes in cellular metabolism, movement from one place to another, etc. There are two major subfamilies of G proteins:

·        The heterotrimeric, or large, G proteins that we will discuss in this lecture.

·        The monomeric, or small, G proteins that are members of the Ras family which will be discussed in a separate lecture. 

G proteins essentially act as molecular “switches” that help to regulate cellular responses initiated through cell surface receptors, such as rhodopsin and the bAR. The heterotrimeric class of G proteins possess three subunits, designated a, b, and g, that display specific biochemical properties: 

a subunit:

·        Binds guanine nucleotide (GDP & GTP).

·        Interacts with specific receptors and effector molecules.

·        Contains an intrinsic GTPase activity that hydrolyzes GTP to form GDP, shutting off the G protein. 

bg subunits:

·        Form a tightly associated complex and are often thought of as a single functional unit.

·        Suppress GDP dissociation, which keeps the a subunit in its inactive, GDP-bound state.

·        Required for high-affinity coupling of the G protein with the receptor.

·        Can regulate the activity of certain effector proteins.

·        Bind to cell membranes by means of lipid modification of the g subunit.

Details of G protein a subunit structure. 

Elucidation of G protein a subunit (Ga) structure and function has come largely from studies involving site-directed and deletional mutatgenesis, studies of naturally occurring mutants (we shall consider one class of Gas mutation, gsp, later in this lecture), construction of chimeric a subunits, biophysical techniques (X-ray crystallography), and use of synthetic peptides. The figure below represents a composite of numerous studies that detailed the various functional domains of several different Ga.

 

The Ga subunit is thought to fold back on itself, assuming a conformation enabling the N- and C-termini to contact one another. Furthermore, these domains face the plasma membrane. From the figure below, several functional domains warrant a brief description (note: refer to the key below the figure to find these domains): 

·        Regions of interaction with receptors & effectors: These areas form the contact points between the receptor, effector and the Ga subunit. These binding sites are spread throughout the linear sequence, but are close together in the three-dimensional structure of the protein. Note that receptor and effector binding domains are among the more diverse regions among a subunits, which gives the different Ga proteins a certain amount of selectivity in terms of which receptors and effectors couple to a particular G protein.

·        Regions of interaction with Gbg: these areas are where the a subunit and bg dimer physically associate. Remember, this association is vital in suppressing release of GDP and maintenance of high-affinity binding to the receptor. Note that one of the regions responsible for heterotrimer association is near the N-terminus of the a subunit, which also partially overlaps the region where receptor–Ga interactions are thought to occur. Thus, an activated receptor may stimulate a G protein by promoting release of the bg dimer through perturbation of a–bg interactions at this point.

·        Switch regions: these are relatively flexible sites within the Ga polypeptide that change the orientation of the effector and G domains (see below), and hence the overall conformation of the a subunit. This change in conformation is driven by the GTP-for-GDP exchange reaction. The result of this altered conformation is release of Ga from bg and exposure of the effector domains to allow for Ga–effector interaction and effector activation.

·        G domains: there are five so-called G domains (G1 through G5) that serve as guanine nucleotide binding sites. The intrinsic GTPase function of the protein also encompasses these regions.

·        Sites of bacterial toxin binding: a valuable experimental tool is the use of specific bacterial toxins, namely cholera (CTX) and pertussis toxin (PTX), that bind to select Ga subunit subtypes. For instance, CTX modifies a key arginine residue within the GTPase of Gas through ADP-ribosylation that blocks GTP-hydrolytic activity. The effect is constitutive activation of Gas and chronic elevation of cAMP levels in the cell. This is what leads to severe diarrhea since elevated cAMP results in water loss through the epitheleal cells lining the intestine. PTX, does not bind to Gas but, instead, binds to the a subunit of Gi (the G protein activated by a2-adrenergic receptors, remember?). Although the mechanism is different with PTX action versus CTX, the net effect is the same. PTX blocks activation of Gi through modification of a cysteine residue near the C-terminus of Gai. When Gi activation is blocked the inhibitory effect on Gs is removed and Gs activity increases, resulting in enhanced cAMP levels. Thus, for both toxins the end effect is increased cAMP production. 

Both CTX and PTX have been used, experimentally, to determine the subclass of G protein to which a GPCR may couple. If you wanted to characterize a newly discovered receptor, for example, one thing you might want to do is to pretreat cells that express the particular receptor with PTX and then expose the receptor to its ligand. If your receptor couples to Gi then you would anticipate that receptor activation of Gi (and, hence, its downstream effector) would be blocked. Depending on the receptor, Gi, in addition to blocking adenylate cyclase, also activates the phosphoinositide pathway so you could measure this effect by assaying for increases in inositol phosphate production. 

X-ray crystallography is a powerful technique that has yielded tremendous insight into the three dimensional structure of proteins. The crystal structures for the a subunits of transducin and Gi were recently solved. Early on, however, much of our understanding about G protein a subunit structure and function came from studies of the closely related small G protein, Ras.

 

      During G protein activation, the resultant conformational changes in the a subunit are reflected in altered orientations of the switch regions. The ability of Ras and Ga to change their conformation in such a manner provides the molecular basis for activating downstream signalling proteins in a regulated fashion. Furthermore, analysis of Ras, Gai, Gat, and elongation factor Tu (a GTP binding protein involved in protein translation) point to a highly conserved activation mechanism shared among proteins that bind guanine nucleotides. The figure below shows the proposed guanine nucleotide-dependent conformational changes in Ras. Note the use of a so-called non-hydrolyzable GTP analog, GppCp, which “locks” the G protein in its active conformation and the differences in switch region orientation versus GDP-bound Ras.

gsp mutations and cancers of the thyroid & pituitary glands

Nature has often provided for us, from alterations in the amino acid sequence of a signalling protein, clues about how structure determines function. Data collected within the last decade have clearly shown that G proteins can participate in mitogenic (growth stimulating) responses and cell transformation (oncogenic potential). In fact, both loss- and gain-of-function mutations in the a subunits for both Gs and Gi have been documented in the literature and are present in several types of tumors. 

One class of genetic alteration, called gsp mutations, occurs in Gas and involve the replacement of either Arg201 with a cysteine residue (R201C) or histidine (R201H) or Gln227 with an arginine (Q227R) or leucine (Q227L). The mutations result in constitutive activation of the a subunit through inhibition of the intrinsic GTPase function of the protein. Thus, the G protein remains in an active, GTP-bound state. gsp mutations affect cells of the pituitary (called pituitary somatotrophs) and thyroid glands. 

Two hormones that bind to distinct GPCRs found on these two cell types, growth hormone-releasing hormone (GHRH) and thyroid stimulating hormone (TSH), elicit increases in intracellular cAMP levels (thus, implicating adenylate cyclase as the effector molecule in this system).  

cAMP can have differing effects on cell proliferation depending on the type of cell one is talking about. In many cell types, cAMP is associated with a growth-suppressive phenotype. However, in pituitary somatotrophs and thyroid cells cAMP has a mitogenic or positive growth effect which is due to the activation of growth promoting genes. It is this positive growth response of the Gs–adenylate cyclase–cAMP signalling pathway that, when a gsp mutation is introduced, can give rise to uncontrolled cell growth and, eventually, tumor formation.  

It is noteworthy that gsp mutations occur in 40% of growth hormone-secreting pituitary adenomas and 30% of thyroid hyperfunctioning adenomas.  

The proliferative effects induced by cAMP in the cell types described above are linked to the activation of the cAMP-dependent kinase (PKA) and its ability to phosphorylate a transcription factor called cAMP responsive element binding protein (CREB). Phosphorylated CREB is then able to promote the transcription of growth activating genes such as the transcription factors c-fos and c-jun as well as a pituitary-specific transcription factor, pit-1. 

 Interestingly, the gsp mutations appear to cause an increase in the degradation of the mutated as subunit (see the Western blot in fig.2 below). Despite the significantly reduced levels of the as–R201 or Q227 mutants, the constitutive activity and resultant growth-promoting phenotype of these proteins predominate.

 

 

Additionally, in both adenoma tumor cells and in cell lines transfected with gsp mutant cDNAs, there appears to be a compensatory mechanism which is activated in response to the elevated levels of cAMP. A cAMP phosphodiesterase (PDE; see fig.1 below) that destroys cAMP by converting it to AMP and is expressed at higher levels in gsp-containing cells can effectively reduce intracellular concentrations of this second messenger. However, only in certain cell types is this mechanism effective in turning off the cAMP-mediate growth response. In particular, cells of the pituitary and thyroid glands do not appear to benefit from hyperexpression of the PDE.