By Mark M. Rasenick, Ph.D.

Departments of Physiology & Biophysics and Psychiatry,

U. Illinois College of Medicine, Chicago, IL 60612

            Receptors for a variety of hormones and neurotransmitters are coupled to their intracellular effectors via G proteins.  G protein mediated signaling systems include stimulation (G_s) and inhibition (G_i) of adenylyl cyclase, the gating of K⁺ (G_i) and Ca²⁺ (G_o) channels, the activation of phosphoinositide phospholipase c (G_q/G₁₁) and Na/H⁺ exchange (G₁₃).  G proteins (hereinafter known as G) are heterotrimeric in structure, and consist of a, b and g subunits, a separating from bg (in solution) subsequent to its binding of GTP.  Molecular weights for the a subunits range from 39-52 kDa, the b's are about 36 kDa and the g subunits range from 8-14 kDa.  Although it is the a subunit which binds GTP, both a and bg subunits engage various G protein effectors.

TABLE I

BIOCHEMICAL CHARACTERISTICS OF HETEROTRIMERIC G PROTEIN FAMILIES ^a

^a Abbreviations are a2-AR: a2-adrenoceptor;  AC: adenylyl cyclase;  Ang-R: angiotensin receptor; Brady-R: bradykinin receptor;  cAMP: cyclic AMP;  DAG: diacylglycerol;  Dopam-R: dopamine receptor;  G_olf: olfactory G protein;   His-R: histamine receptor;  Musc-R: muscarinic receptor;  para-H-R: parathyroid hormone receptor;  PDE: phosphodiesterase;  PGE1-R: prostaglandin E1-R;  PLC: phospholipase C; PLA: phospholipaseA; Rho: rhodopsin ; 5-HT-R: serotonin receptor;  Throm-A2-R: thromboxane A2 receptor;  Vaso-R: vasopressin receptor.  The receptor list is not meant to be inclusive.

Receptor-G protein coupling

            As is evident from Table 1, a given agonist might activate a large number of related receptors and those receptors may well be coupled to disparate intracellular effectors.  Serotonin, for example, is known to activate and inhibit adenylyl cyclase, activate phospholipase and inhibit voltage sensitive Ca²⁺ channels.  In order to do this, a number of 5HT receptor subtypes employ a variety of G proteins (G_s, G_i, G_q/11 and G_o) to engage those effectors.  Determining the precise G protein coupled to an effector might be critical for developing the ability to design molecular therapeutics.  For example, if a receptor is coupled to multiple intracellular processes, the design of a drug (a peptide mimetic) which works at the receptor-G protein interface might be the best way to engineer specificity (Rasenick, et al., 1993) .

            Identification of the G proteins involved is crucial and is often best served by multiple techniques.  For example, it is possible to use an antisense approach, either injected into cells (Kleuss, et al., 1991)  or incorporated into the genome (Liu, et al., 1994) ) to eliminate the expression of individual G protein a subunits involved in transducing the signal from a given receptor.  While such an approach can yield valuable data, several G proteins appear to be capable of assuming the role of a missing compatriot.  Such plasticity is confounding to antisense and knockout studies.

            Another approach to the identification of G proteins involves photoaffinity labeling with AAGTP.  Such techniques allow precise identification of the G protein activated as the direct result of agonist occupancy of a receptor.   Furthermore, if an antagonist of a given form of a receptor was present during photoaffinity labeling studies, an investigator might identify which of several G proteins activated by a given agonist is the "right one" for a given process.

 Properties of AAGTP

            AAGTP is a hydrolysis-resistant GTP analog.  As such, the radioactive phosphate can occupy either the alpha or gamma position without fear of hydrolysis or phosphorylation.  This can be controlled for by running a separate incubation which is not subject to UV irradiation.  

            It should not be anticipated that all proteins binding AAGTP will become substrates for covalent attachment.  The efficiency of covalent photoincorporation is about 2.5% (Gordon and Rasenick, 1988)  which is normal for most aryl-azide compounds.  It is noteworthy, however, that this will afford a level of detectability of G protein which exceeds that of any SDS-PAGE protein stain.

            While indirect room lighting will photolyze AAGTP very slowly, it is a good idea to protect from bright fluorescent light or sunlight.  Incandescent lighting will provide a good working environment.  If practical, gold fluorescent lights (e.g. Phillips F40 GO) are ideal.  

AAGTP labeling of membrane G proteins:  Agonist independent labeling

            It is likely that G proteins in membranes are subjected to a multitude of regulatory factors.  In order to examine the behavior of specific proteins within this complex milieu, AAGTP serves as the most useful tool.

            Labeling of membranes from rat cerebral cortex is accomplished by incubating the membranes with 0.1 µM [³²P] AAGTP for 3 minutes at 23°C, followed by centrifugation -washing (15,000xg) twice with 20 volumes of buffer.  Such treatments allow only tightly bound AAGTP to remain associated with membranes prior to photolabeling (Hatta, et al., 1986) .  In synaptic membranes, receptors are uncoupled from G proteins subsequent to the disruption of cells.  Thus, even in the absence of an agonist, at least four G proteins in rat cerebral cortex synaptic membranes bind AAGTP  (Gordon and Rasenick, 1988) .

Agonist-induced AAGTP binding: Studies of receptor-coupled events

For G proteins linked to receptors, binding of AAGTP in the absence of agonist is significantly lower (Offermanns, et al., 1991; Offermans, et al., 1990) , but there is significant background binding which must be eliminated in order to see clear agonist dependence of AAGTP labeling.  Background binding of AAGTP can be reduced considerably by the inclusion of 1 mM mercaptoethanol in the buffer.  This is advisable in the presence or absence of agonist.   Another technique which reduces background AAGTP binding while increasing agonist-specific AAGTP incorporation involves inclusion of GDP  in the binding buffer (Offermanns, et al., 1991).  Although, 10-30 mM GDP was found to reduce the overall incorporation of AAGTP by 39-41 KDa proteins by 50 to 100 fold, Offermans et.al. (Offermanns, et al., 1991)  were able to show a 2 fold increase in labeling induced by agonist in membranes from NG108-15 cells.  In other systems, the amount of GDP required to increase agonist specific labeling could be much lower.  Further, GDP seems to be more critical to see hormone dependence of AAGTP binding to G proteins of the ai/o or the aq/11 families. 

Photoaffinity labeling of membrane G proteins:  Stepwise approach

1)  Membranes are prepared and added to 2 ml screw-top microfuge tubes.  Total assay volumes should be kept to 30-50 ml, but can go as high as 100 ml.  The total amount of membranes will be dependent upon the identification step.  If SDS-PAGE and autoradiography are planned, 30 mg of membrane protein/gel lane are probably sufficient.  If a western blot is planned, 90 mg are better and if immunoprecipitation of the sample will be done, use 150 mg.  For photolabeling 10 mM Hepes pH 7.4,  1 mM MgCl₂, 1 mM b-mercaptoethanol is a good universal buffer to use.

2)  If agonist-dependent labeling is planned, membranes should be incubated with GDP (amount determined experimentally) for 5-10 minutes at 23°C.  Following this, ³²P AAGTP is added.  If agonist is not present, 0.1 mM is a good concentration to use.  When doing agonist dependent labeling, AAGTP concentrations of 1 nM will give diminished overall labeling, but are likely to show increased agonist specificity.

3)   After 3 minutes incubation with AAGTP, agonist is added and incubation continued for 10 minutes (this time will also need to be determined experimentally).

4)  Following incubation, tubes are places in a metal block on ice and subjected to UV irradiation (254 nm; 9W manufacturer’s rating, 1-6 mW/cm² determined by light meter) with a hand-held lamp held at 4 cm for 3 minutes.  

5)  Photoincorporation is stopped by adding ditihiothreitol to 10 mM.  Membranes can then be washed by the addition of 1 ml labeling buffer and centrifugation in a microcentrifuge.  Pellets are resuspended in sample buffer and subjected to SDS PAGE.  Alternately, concentrated (3X) sample buffer is added to the solution and the sample is applied to SDS PAGE.  Gels can either be dried and exposed to film or phosphorimaging, or proteins from wet gels can be transferred to nitrocellulose and processed for immunoblot analysis.

Immunoblot analysis of G proteins labeled by AAGTP

 If immunoblots are prepared, color development (as opposed to chemiluminescence detection) is preferred, since autoradiography of the nitrocellulose can be compared to the bands corresponding to a given G protein a subunit.  If radiographic analysis is performed in a phosphorimager, a printout should be made on a transperancy so that a definitive analysis can be made of the G protein species which has bound AAGTP.

Since most G proteins are clustered between 40 and 45 kDa, good separation is often difficult.  This can sometimes be achieved by lowering the bisacrylamide concentration (prepare a stock of 30/0.4 % acrylamide/bisacrylamide) and adding 8 M urea to the resolving gel (See figure 1 on page 9). 

Immunoprecipitation of AAGTP labeled G proteins

Absolute identification of AAGTP-labeled G proteins is dependent upon their specific immunoprecipitation.  This is accomplished by rendering G proteins from the AAGTP-labeled membrane pellet soluble with lubrol or octyl-glucoside (triton x-100 or triton x-114 will not solubilize all membrane G proteins, Yan, et al., 1996).  Following immunoprecipitation by standard methods, SDS PAGE is performed and G proteins are identified.

Troubleshooting:

1) Failure to observe agonist-induced AAGTP labeling

This is a common problem with a complex solution.  The concentrations of GDP, Mg²⁺ and AAGTP may well be different for each agonist in each tissue.  Different G proteins coupled to the same receptor may also show distinct properties in this regard, so labeling profiles for a single agonist should be compared under different GDP, Mg²⁺ and AAGTP concentrations to be sure that more than one G protein is not activated (albeit under different conditions).  It is noteworthy that receptor coupling is often lost following cell disruption.  This is especially true for various agonists in a number of brain regions (Rasenick, et al., 1989; Rasenick, et al., 1993) .  One solution is to use a permeable cell system, where tight coupling between receptor and G protein is maintained (Rasenick, et al., 1993; Rasenick, et al., 1994) .

References:

Bernstein, G., L., B.J., Jhon, D.-Y., Exton, J.H., Rhee, S.G. and Ross, E.M., (1992) phospholipase C-beta 1 is a GTPase-Activating Protein for G_q/11, Its Physiologic Regulator. Cell. 70: 411-418

Gordon, J.H. and Rasenick, M.M., (1988) In situ binding of a photo-affinity GTP analog to synaptic membrane G-proteins. FEBS Lett. 235: 201-206

Hatta, S., Marcus, M.M. and Rasenick, M.M., (1986) Exchange of guanine nucleotide between GTP-binding proteins that regulate neuronal adenylate cyclase. Proc. Natl. Acad. Sci., USA. 83: 5439-5443

Kleuss, C., Heschler, J., Ewel, C., Rosenthal, W., Schultz, G. and Wittig, B., (1991) Assignment of G-protein subtypes to specific receptors inducing inhibition of calcium currents. Nature. 353: 43-48

Liu, Y.F., Jakobs, K.H., Rasenick, M.M. and Albert, P.R. (1994) G protein specificity in receptor-effector coupling: Analysis of the roles of G_o and G_i2 in GH4C1 pituitary cells. J. Biol. Chem. 269: 13880-13886

Offermanns, S., Schultz, G. and Rosenthal, W., (1991) Evidence for Opioid Receptor mediated Activation of the G-proteins, G_o and G_i2 in Membranes of Neuroblastoma X Glioma (Ng108-15) Hybrid Cells. J. Biol. Chem. 266: 3365-3368

Offermans, S., Schafer, R., Hoffmann, B., Bombien, E., Spicher, K., Hinsch, K.-D., Schultz, G.and Rosenthal, W., (1990) Agonist-sensitive binding of photoreactive GTP analog to a G-protein alpha-subunit in membranes of HL-60 cells. J. Biol. Chem. 260: 14-18

Rasenick, M.M., Hughes, J.M. and Wang, N., (1989) Guanosine-5'-O-thiodiphosphate functions as a partial agonist for the receptor-independent stimulation of neural adenylate cyclase. Brain Research. 488: 105-113

Rasenick, M.M., Lazarevic, M., Watanabe, M. and Hamm, H.E., (1993) Permeable Cell Systems as Models for Studying  Disruption, by Site-Specific Synthetic Peptides, of Receptor-G protein-Effector Coupling. Methods. 5: 252-257

Rasenick, M.M., Watanabe, M., Lazarevic, M.B., Hatta, S. and Hamm, H.E., (1994)  Synthetic peptides as probes for G protein function:  Carboxyl terminal Gas peptides mimic G_s and evoke high affinity agonist binding to b-adrenergic receptors. J. Biol. Chem. 269: 21519-21525

Yan, K., Greene, E., Belga, F. and Rasenick, M.M., (1996) Synaptic membrane G proteins are complexed with tubulin in situ. J. Neurochem. 66: 1489-1495

Mark M. Rasenick Professor Physiology and Biophysics, and Psychiatry Director, Biomedical Neuroscience Training Program Cytoskeletal modulation of neuronal signal transduction http://www.uic.edu/depts/mcpb/pages/rasenick/rasenick.htm

Most of the work in my laboratory involves the possibility that a structural protein, tubulin, conveys a dynamic response and alters the ability of a neurotransmitter to convey it's message across the synapse. It appears that tubulin forms specific complexes with certain G proteins (the intracellular harbingers of neurotransmitter messages) and activates them via the direct transfer of GTP. In this manner, a neurotransmitter which had no direct effect upon a certain G protein mediated pathway could affect that pathway by increasing (or decreasing) the activation of a G protein via tubulin. It is also possible that changes in synaptic shape or the synaptic cytoskeleton, which occur during chronic neural activity, might alter responsiveness to a neurotransmitter. Several changes in the cytoskeleton have been noticed in both the developing and the degenerating nervous system. The neurotransmitter G protein axis could be responsible for altered cytoskeletal form (and consequently, altered synaptic development) and might provide an forum whereby activity in one neuron could dictate development of the synapse. A variety of molecular techniques are being employed to elucidate this system. It is hoped that a better understanding of the interface between the cytoskeleton and G protein mediated signal transduction systems could provide a better understanding of brain and mind function and dysfunction.

Rasenick, M.M.  Gs (a poem).  Trends in Biochem. Sci. 17:71, 1992.

Biographical Sketch

Selected Publications:

1)  Popova, J.S., Garrison, J.C., Rhee, S.G. and RASENICK, M.M. Tubulin, Gq and phosphatidylinositol 4,5-bisphosphate interact to regulate phospholipase C 1 signaling. J. Biol. Chem. 272:6760-6765, 1997.

2)  Roychowdhury, S. and RASENICK, M.M. G-protein bg subnits regulate microtubule assembly.  J. Biol. Chem. 50:31576-31581, 1997.

3)  Roychowdhury, S., Panda, D., Wilson, L. and RASENICK, M.M.   G protein a subunits activate tubulin GTPase and modulate microtubule polymerization dynamics.  J. Biol. Chem. 274:13485-13490, 1999.

4)  Toki, S., Donati, R.J. and RASENICK, M.M.  Treatment of C6 Glioma cells and rats with antidepressant drugs increases the detergent extraction of Gsa from plasma membrane.  J. Neurochem. 73:1114-1120, 1999.

5)  Popova, J.S. and RASENICK, M.M.  Muscarinic receptor activation promotes the membrane association of tublin for the regulation of Gq-mediated phospholipase Cb1 signaling.  J. Neurosci. 20:2774-2782, 2000.

6)  Yan, K., Popova, J.S., Moss, A., Shah, B. and RASENICK, M.M.  Tubulin stimulates adenylyl cyclase activity in C6 glioma cells by bypassing the b-adrenergic receptor: a potential mechanism of G protein activation.  J. Neurochem., 76:182-190, 2001.

7)  Donati, R., Thukral, C. and RASENICK, M.M.  Chronic treatment of C6 glioma cells with antidepressant drugs results in a redistribution of Gsa.  Mol. Pharmacol., 59(6):1-7, 2001.

8)  Yu, J.Z. and RASENICK, M.M.  Real-time visualization of a fluorescent Gas: Dissociation of the activated G protein from the plasma membrane.  Mol. Pharmacol., 61:352-359, 2002.

9)  Yu, J.Z. and RASENICK, M.M.  Transient expression of fluorescent tau proteins promotes process formation in PC12 cells:  Contributions of the tau C-terminus to this process.  J. Neurosci. Res., 67:625-633, 2002.

10) Popova, J.S., Greene, A., Wang, J. and RASENICK, M.M.  Phophatidylinositol 4,5-bisphosphate modifies tubulin participation in PLCb1 signaling. J. Neuroscience, 22:1668-1678, 2002. 

mailto: Dr. Rasenick

Return to Physiology Homepage

Biomedical Neuroscience Training Program