microscope. For the electron microscopy analysis, animals were fixed with glutaraldehyde and osmium tetraoxide (34). Five or six L4 or young adult animals were aligned within a small agar block, embedded, and sectioned together. Sections were poststained with uranyl acetate and lead citrate. 4. R. Durbin, thesis. University of Cambridge, Cambridge, England (1987). 5. C. Garriga, C. Desai, H. R. Horvitz. Development 117, 1071 (1993. 6. S. Kim and W. G. Wadsworth, data not shown. 7. The nid-7(RNAi) animals were generated as described previously [A. Fire et al., Nature 391. 806 (1998)l by using a 1-kb sequence from exon 8, which was cloned into pBluescript (Stratagene) as template for RNA synthesis. RNA was produced by both T3 and T7 RNA polymerase, and the reactions were pooled before being injected into the intestines of edls2O(unc-779::CFP) transgenic animals. Phenotypes of the nervous system were observed under epifluorescence microscopy. 8. Seven PCR fragments, including the whole coding sequence and intron region, were amplified from the genomic DNA of nid-l(ur47) animals. PCR fragments were cloned into a pBluescript vector and subsequently sequenced by automatic sequencer. The mutation was confirmed by sequencing two independent PCR fragments. . 16. lmmunostaining was performed by using freeze-fracture and methanol-acetone fixation as previously described (35). Polyclonal antibodies raised against mouse nidogen and a mouse monoclonal antibody against myosin heavy chain B (UNC-54) were used. For costaining, anti-rabbit fluorescein-conjugated and anti-mouse rhodamine-conjugated secondary antibodies were used. . 24. Transgenic strains were generated by standard methods (36). plM#194, an expression construct for nid-7, was constructed by cloning the 2.5-kb 5' flanking region of nid-7 into pPD 95.77 vector (from A. Fire). This CFP construct was coinjected at 10 pg/ml with pRF4 at 100 pg/ml. To establish a stable line, lM329 urls757 [plM#194, pRF4], transgenes were integrated by y-ray irradiation. For the ectopic expression construct of nid-7, constructs plM#195, plM#196, and plM#197, were made by using the 7-kb genomic nid-7 coding region, which was amplified by high-fidelity PCR, ligated to Nhe I-Bgl Il-digested vectors, pPD96.41, pPD49.83, and pPD96.52 (from A. Fire). These vectors contained 5' flanking regulatory sequences of mec-7, hsp76-47, and myo-3, respectively (36). The unc-779 regulatory sequence was amplified by using plM175 as template (23), and cloned into the pPD49.26 vector (from A. Fire) to construct plM#198. These constructs were injected at 10 pg/ml, with pRF4 into nid-l(ur47); kyls723 (zc27::CFP) animals. The resulting strains are IM330 urEx752 [plM#195]; nid-7 (ur47); kys123(zcZI::GFP); IM331 urEx753 [plM#196]; nid-l(ur47); kyls723 (zc27::tFP). lM332 urEx754 [plM#197]; nid-l(ur47); kyls723(zc27::CFP), 11' 1333 urEx755 [plM#198]; nidl(ur47); kyIs723(zc27::CFP). Ectopic expression of nid-7 was checked by in situ hybridization. IM331 emblyos collected 1 to 6 hours after being laid were heat-shock...

The existence of receptor dimers has been proposed for several G protein-coupled receptors. However, the question of whether G protein-coupled receptor dimers are necessary for activating or modulating normal receptor function is unclear. We address this question with somatostatin receptors (SSTRs) of which there are five distinct subtypes. By using transfected mutant and wild type receptors, as well as endogenous receptors, we provide pharmacological, biochemical, and physical evidence, based on fluorescence resonance energy transfer analysis, that activation by ligand induces SSTR dimerization, both homo-and heterodimerization with other members of the SSTR family, and that dimerization alters the functional properties of the receptor such as ligand binding affinity and agonist-induced receptor internalization and up-regulation. Double label confocal fluorescence microscopy showed that when SSTR1 and SSTR5 subtypes were coexpressed in Chinese hamster ovary-K1 cells and treated with agonist they underwent internalization and were colocalized in cytoplasmic vesicles. SSTR5 formed heterodimers with SSTR1 but not with SSTR4 suggesting that heterodimerization is a specific process that is restricted to some but not all receptor subtype combinations. Direct protein interaction between different members of the SSTR subfamily defines a new level of molecular cross-talk between subtypes of the SSTR and possibly related receptor families.

No abstract

Heptahelical receptors (HHRs) are generally thought to function as monomeric entities. Several HHRs such as somatostatin receptors (SSTRs), however, form homo-and heterooligomers when activated by ligand binding. By using dual fluorescent ligands simultaneously applied to live cells monotransfected with SSTR5 (R5) or SSTR1 (R1), or cotransfected with R5 and R1, we have analyzed the ligand receptor stoichiometry and aggregation states for the three receptor systems by fluorescence resonance energy transfer and fluorescence correlation spectroscopy. Both homo-and heterooligomeric receptors are occupied by two ligand molecules. We find that monomeric, homooligomeric, and heterooligomeric receptor species occur in the same cell cotransfected with two SSTRs, and that oligomerization of SSTRs is regulated by ligand binding by a selective process that is restricted to some (R5) but not other (R1) SSTR subtypes. We propose that induction by ligand of different oligomeric states of SSTRs represents a unique mechanism for generating signaling specificity not only within the SSTR family but more generally in the HHR family. H eptahelical receptors (HHRs) constitute the largest single family of transmembrane signaling molecules that respond to diverse external stimuli such as hormones, neurotransmitters, chemoattractants, odorants, and photons. Although these receptors have been generally thought to function as monomeric entities, there is growing evidence that a number of HHRs assemble as functional homo-and heterooligomers (1, 2). Dimerization seems to be necessary for function of the class C subfamily of HHRs comprising the metabotropic glutamate, calcium sensing, the GABAB, and pheromone receptors that are targeted to the plasma membrane as preformed dimers which are stabilized by ligand binding (3-7). Several HHRs such as somatostatin receptors (SSTRs), dopamine receptors, gonadotrophin-releasing hormone receptor (GnRHR), luteinizing hormone͞chorionic gonadotrophin hormone receptor, and chemokine receptors, however, which belong to the rhodopsin-like class A subfamily of HHRs, assemble on the membrane as homo-and heterooligomeric species in response to agonist activation (8-15).In the case of SSTRs, we have shown by photobleaching fluorescence resonance energy transfer (pbFRET) that the human (h) type 5 receptor (hSSTR5 or R5) exists in the basal state as a monomer, and that activation by ligand induces dose-dependent oligomerization (8). When coexpressed with another SSTR (hSSTR1 or R1) or an unrelated HHR such as the dopamine 2 receptor (D 2 R), R5 also forms a heterooligomer that displays pharmacological properties distinct from those of either of the separate receptors (9). Little is known about the stoichiometry of ligand-receptor reactions or the specificity for homoand heterooligomeric interactions between two receptors that are coexpressed in the same cell. By using dual fluorescent ligands simultaneously applied to live cells monotransfected with R5 or R1, or cotransfected with R5 and R1, we have analyzed the ...