Localization of G protein-coupled receptor kinases (GRKs) in the avian retina

Dopamine receptor D4 internalization requires a beta-arrestin and a visual arrestin

PURPOSE

The G-protein coupled receptor (GPCR) Dopamine Receptor D4 (DRD4) plays an essential role in cAMP regulation and gap junctional coupling in the photoreceptors, where DRD4 expression is under circadian control. Previous in vitro transfection studies of human DRD4 desensitization have reported that DRD4 is not internalized upon dopamine stimulation when beta-arrestin is co-transfected with DRD4. We hypothesized that the visual arrestins, ARR1 and ARR4, play a modulatory role in DRD4 desensitization in the photoreceptors.

METHODS

To test this hypothesis, immunohistochemistry analysis of mouse retinas was used to determine the cellular localization of beta-arrestins and DRD4 in photoreceptors. In vitro studies were performed in HEK293T cells transiently transfected with human DRD4 and arrestins. First, co-immunoprecipitation experiments were executed to test protein-protein interactions and to investigate the effect of dopamine stimulation. Second, immunohistochemistry analysis was implemented to study DRD4 internalization and translocation of ARR4.

RESULTS

Immunohistochemistry studies of mouse retinas confirmed the expression of beta-arrestin 2, ARR1 and ARR4, as well as DRD4 in mouse cone photoreceptor inner segments. Co-immunoprecipitation experiments revealed a dopamine-dependent protein-protein interaction between human DRD4 and ARR4. In vitro internalization experiments showed that no detectable internalization of DRD4 was observed with any single arrestin co-transfected. However, a dopamine-dependent internalization of DRD4 was observed with three out of six sets of two arrestins co-transfected with DRD4. Each of these pairs of arrestins contained one visual arrestin and one beta-arrestin, and no internalization was observed with either two visual arrestins or two beta-arrestins. Additional time-course experiments revealed that in vitro, ARR4 translocates to co-localize with DRD4 at the plasma membrane in response to 30min of dopamine stimulation.

CONCLUSIONS

The results have functional implications and we hypothesize that the desensitization and internalization of DRD4 in photoreceptors are synergistically mediated by both visual and beta-arrestins. These results are additionally unique because they demonstrate for the first time that at least one G-protein coupled receptor, DRD4, requires two arrestins for desensitization and internalization, and opens up the possibility that other G-protein coupled receptors may require more than one arrestin for desensitization and/or internalization.

Modulation of the neuronal response to ischaemia by somatostatin analogues in wild-type and knock-out mouse retinas

Somatostatin acts at five G protein-coupled receptors, sst(1)-sst(5). In mouse ischaemic retinas, the over-expression of sst(2) (as in sst(1) knock-out mice) results in the reduction of cell death and glutamate release. In this study, we reported that, in wild-type retinas, somatostatin, the multireceptor ligand pasireotide and the sst(2) agonist octreotide decreased ischaemia-induced cell death and that octreotide also decreased glutamate release. In contrast, cell death was increased by blocking sst(2) with cyanamide. In sst(2) over-expressing ischaemic retinas, somatostatin analogues increased cell death, and octreotide also increased glutamate release. To explain this reversal of the anti-ischaemic effect of somatostatin agonists in the presence of sst(2) over-expression, we tested sst(2) desensitisation because of internalisation or altered receptor function. We observed that (i) sst(2) was not internalised, (ii) among G protein-coupled receptor kinases (GRKs) and regulators of G protein signalling (RGSs), GRK1 and RGS1 expression increased following ischaemia, (iii) both GRK1 and RGS1 were down-regulated by octreotide in wild-type ischaemic retinas, (iv) octreotide down-regulated GRK1 but not RGS1 in sst(2) over-expressing ischaemic retinas. These results demonstrate that sst(2) activation protects against retinal ischaemia. However, in the presence of sst(2) over-expression sst(2) is functionally desensitised by agonists, possibly because of sustained RGS1 levels.