GPCR Dimers Are Not Universally Required as Prerequisites for G Protein Activation

Importantly, though, these three examples are from the G branch of GPCRs (according to the GRAFS nomenclature [5]), which comprises the meta-

botropic glutamate (mGluR), calcium sensing (CaSR), and GABA B receptors. The relevance of dimerization for G protein activation is much less clear, if the other GPCR subfamilies are examined. This is, in particular, true for the largest, that is, the R branch of rhodopsin-l ike receptors. The controversy is most readily evident with the eponymous member of the branch, namely rho-dopsin itself. Rhodopsin dimers have been visualized in crystals [54] and observed by atomic force microscopy in native disk membranes [55], but it is not clear if they are meaningful for understanding the mechanism of G protein activation:

1. Under fully dark adapted conditions, a single photon—captured by a single chromophore in a single rhodopsin—is likely to activate a rod to elicit the primary visual response. Given this exquisite sensitivity and the powerful amplification, it is a priori difficult to conceptualize a role of dimeric rhodopsin, unless the second rhodopsin in the dimer need not be active. This model, however, implies precoupling, that is, prebinding of transducin Gt to (inactive dark) adapted rhodopsin. This is inconceivable because rhodopsin is in (up to ~12-fold) excess over transducin; precoupling would preclude the catalytic action of rhodopsin required for signal amplification and make it impossible for rhodopsin to operate under the range of illuminations encountered during a given diurnal cycle [56] .

2. Although rhodopsin may assemble to dimers and higher-order arrays in rod outer segments [55], this does not a priori prove that it is the dimeric form that activates the G protein. In fact, the dimer/monomer discussion has not only inspired insightful reviews [ 57], but has also stimulated ingenuous experiments designed to test the catalytic prowess of monomeric rhodopsin: if solubilized in the detergent dodecyl maltoside, rho-dopsin fulfills all criteria of a monomerically dispersed protein. In this monomeric form, rhodopsin activates transducin with catalytic perfection, that is, in a diffusion-limited manner [58] . The rate of activation in solution is lower than that observed in disc membranes but this can be accounted for by the restricted diffusional freedom/directed orientation imposed by the membrane rather than any impairment of monomeric rhodopsin. This interpretation is supported by observations on rhodopsin inserted into nanodiscs of defined size [59] and high-density lipoprotein (HDL) particles of ~10nm [60] . Under these conditions, the lipid particles are either too small to contain more than one rhodopsin [60] or their size can be defined to allow for insertion of one or two rhodopsin molecules -61] . The results are unequivocal: a single rhodopsin molecule suffices to efficiently activate transducin. In fact, when inserted into the nanodisc, the second rhodopsin does not enhance catalysis: the turnover number is halved, indicating that it does not contribute to any appreciable extent to transducin activation [61]. Similarly, transducin only stabilizes one rhodopsin molecule in the metarhodopsin II (MII) state

(the equivalent of the ternary HRG complex). These data may be interpreted to justify a dimeric arrangement as outlined in Fig. 4.2b), with one active MII state stabilized by Gt and the second rhodopsin as an inactive scaffold. However, the interpretation of the experiment is limited by the fact that the orientation of rhodopsin dimer is not known. If it is random, there are only 50% dimers with parallel orientation and only these dimers can bind transducin. In fact, earlier experiments showed that—at saturating G protein levels—the number of receptors trapped as HRG complexes in reconstituted vesicles approached only 50% of that observed in detergent solutions [62] ; in contrast, in fusion proteins, where Ga is directly tethered to the receptor, virtually all receptors are capable of forming high- affinity complexes [28] . This clearly indicates that receptor orientation can be limiting. Taken together, these data show that monomeric rhodopsin is perfectly capable of activating transducin. An analogous conclusion can also be drawn for the P2-adrenergic receptor, which—when confined to an HDL particle as monomer—activates its cognate G protein G- very efficiently [61]. This is remarkable because fluorescence resonance energy transfer (FRET) microscopy and bioluminescence resonance energy transfer (BRET) recordings have implied that, in living cells, the P2- adrenergic receptors are found in various (homo- and hetero- ) dimeric arrangements (for review, see Reference 46).

At the very least, one would have to concede that—at the current stage—explanations that invoke a universal role of dimers are not parsimonious. The conclusion is justified that receptor dimers (or higherorder oligomers) appear to be dispensable for G protein activation by those rhodopsin branch members, where it has been subjected to rigorous testing.

3. Rhodopsin and its bleached version opsin differ in their sensitivity to thermal denaturation. As mentioned above, a dimeric model (similar to that illustrated in Fig. 4.2b) is conceivable, but it implies a mixed active/ inactive arrangement. Thus, if the state of rhodopsin is sampled by differential scanning, the presence of dimers in rod (outer segment) disk membranes ought to be evident from a change in the thermal denatur-ation curve. This has been examined recently [63]: the data provided little evidence for the presence of dimeric rhodopsin, leading to the conclusion that the bulk of rhodopsin in the disc membrane is a monomer. It is generally accepted that dimeric arrangements visualized by X-ray crystallography only provide circumstantial evidence for the state of the protein in a living cell. In contrast, atomic force microscopy can, in principle, sample the state(s) of a protein in its native environment. Thus, the observations of rhodopsin dimers/higher-order oligomers by atomic force microscopy [55] cannot be easily dismissed, because it must reflect a biologically relevant property of the protein. It has been pointed out that the mica surface (employed in atomic force microscopy) depletes the membranes of lipids [57]. The rhodopsin concentration in the disk membrane is very high (>25,000 molecules/^m2), and the ratio of phospholipid to rhodopsin is about 1:65 -64] - accordingly, the surrounding lipid anulus can only comprise a few layers of phospholipids. Any manipulation, which reduces the solvation of the hydrophobic segment of rhodopsin (including phospholipid depletion and hydrophobic mismatch), is likely to promote oligomeric assemblies -65, 66] - Thus, the monomer/dimer equilibrium may reflect local packing and may be relevant in vivo, when GPCRs are present in high concentrations, for example, disc membranes and synapses. But these dimers or higher-order oligomers are not necessary for G protein activation.

0 0

Post a comment