Tải bản đầy đủ - 0 (trang)
2 Biosynthesis, Export, and Cell-Surface Stabilization

2 Biosynthesis, Export, and Cell-Surface Stabilization

Tải bản đầy đủ - 0trang

196



M.L. Perreault et al.



receptors at the cell surface which is contributed to, in part, by the processes of

receptor biosynthesis and export trafficking.



8.2.1 Biosynthesis and Cell-Surface Trafficking

of Dopamine Receptors

Similar to all GPCRs, the synthesis and folding of dopamine receptors takes place

in the ER. At the completion of this process, the receptors are subjected to a stringent ER quality control system that functions to ensure only correctly configured

receptors can exit the ER and continue their migration through the endoplasmic

compartments toward the Golgi complex. Vital contributors to the ER quality control process are the molecular chaperones, a group of ER resident proteins that have

the dual function of assisting in the overall speed and efficiency of glycoprotein

folding, and inhibiting the export of misfolded proteins from the ER.

8.2.1.1 Calnexin

Calnexin is a chaperone protein with lectin-like activity that recognizes and

binds monoglucosylated N-linked oligosaccharides that have been attached to

GPCRs during the translational process. This interaction not only functions to

prevent the formation of aggregates and promote receptor folding but also serves

to anchor the glycoproteins within the ER until their native conformation is

attained or until they are targeted for degradation. An involvement of calnexin

in the biosynthesis of D1 and D2 receptors has recently been reported [14]. The

association of calnexin with the receptors is mediated, at least in part, by glycosylation as the inhibition of glycosylation, through mutation or with the inhibitor

tunicamycin, diminished calnexin interactions with the receptors. However, the

finding that their association was not completely abolished in the absence of

glycosylation is indicative of a direct interaction also with the receptor protein

[14]. It has been postulated that these glycan-dependent and glycan-independent

actions of calnexin on dopamine receptors may mediate, respectively, the chaperone versus ER retention functions of the protein. This hypothesis was supported

by the finding that while the inhibition of glycosylation restricted the cellsurface expression of the D1 receptor, the increase in calnexin binding that was

observed to a trafficking-impaired D1 receptor was insensitive to glycosylation

inhibitors [14].

Although calnexin has been demonstrated to be involved in the biosynthesis of

GPCRs [15, 16], a role for this chaperone in oligomerization has not been identified. Indeed, it has been shown that calnexin may not even associate with D1 and D2

receptor oligomeric complexes, but only bind to monomeric species [14]. However,

while these findings may appear to exclude a role for calnexin in oligomer assembly, it has been suggested that the ability of calnexin to retain individual receptor

monomers may function to facilitate the formation of oligomeric complexes [14].



8



Regulation of Dopamine Receptor Trafficking and Responsiveness



197



While the mechanisms underlying this hypothesis require further investigation, one

possibility is that calnexin may simply restrain each subunit in a spatial orientation

that makes the monomer more readily accessible for receptor–receptor interactions.

Evidence for self-oligomerization of calnexin [17] suggests a mechanism by which

this chaperone may facilitate these interactions by assisting monomers into close

proximity so as to promote oligomer assembly.



8.2.1.2 The Triple Phenylalanine Export Motif and DRiP78

Export from the ER has been shown to be a critical rate-limiting step in the trafficking of GPCRs to the cell surface [18]. For many GPCRs the selection for ER

export appears to be dependent on the proximal portion of their carboxyl terminus

[19–23] and through the use of site-directed mutagenesis, several motifs within the

carboxyl terminus have been identified that may serve as ER export signals [19, 24–

26]. One such motif has been reported for the D1 receptor [24]. It was established

that the highly conserved triple phenylalanine motif, FxxxFxxxF, was integral for

the cell-surface expression of the D1 receptor as substitution mutations within the

motif resulted in ER retention and loss of ligand binding. Moreover, fusion of the

motif to an intracellularly trapped receptor protein restored its ER export properties

and conferred normal protein transport to the cell surface. Evidence suggests that

the motif may promote vesicular transport from the ER or downstream from the

ER, by interacting with the vesicular coat protein complex COPI [27], a complex

that has been implicated in transport pathways throughout the ER-Golgi network

(reviewed [28]) including exit from the ER [29]. Precipitation assays revealed that

the D1 receptor could associate specifically with the γ-subunit of COPI. Although

such an interaction between carboxyl terminal motifs and coat protein complexes

has not yet been reported for other GPCRs, it has been exhibited for a number of

non-GPCR proteins [30–32].

The export capabilities of the FxxxFxxxF motif can be regulated by the molecular chaperone dopamine receptor interacting protein 78 (DRiP78), a membraneassociated ER resident protein [24]. Transport of the D1 receptor appeared to be

highly sensitive to the intracellular levels of DRiP78 as overexpression of the protein led to ER retention. It has been suggested that DRiP78 may function to mask

the ER export signal thereby preventing the interaction of the motif with another

complex associated with vesicular transport, such as the coatomer complexes, for

example. However, given that DRiP78 sequestration similarly resulted in reduced

D1 receptor cell-surface expression, these results suggested that discrete levels of

the protein may be required for sufficient export trafficking of the D1 receptor from

the ER [24]. It is possible that, in addition to its role in ER retention through motif

binding, DRiP78 may function to assist in protein folding and/or oligomer assembly. Such a role has been recently demonstrated for DRiP78 in the assembly of

G-protein βγ heterodimers [33]. If such was the case for D1 receptors, insufficient

levels of DRiP78 could result in misfolded or unformed oligomer complexes leading

to ER retention and reduced cell-surface expression.



198



M.L. Perreault et al.



8.2.1.3 Role of Glycosylation in Receptor Cell-Surface Targeting

Studies have shown that glycosylation is requisite for the optimal expression of

certain dopamine receptor subtypes at the plasma membrane. It has been previously reported that the loss of N-linked glycosylation by the inhibitor tunicamycin,

or the mutation of a single N-linked glycosylation site (Asp4 ), did not attenuate

the cell-surface expression of the D1 receptor [34]. Yet it has been more recently

reported that tunicamycin, under similar treatment conditions, inhibited the cellsurface expression of the D1 receptor by approximately 27% [14]. Furthermore, this

study demonstrated that a D1 receptor mutant missing both N-linked glycosylation

sites (Asp4 and Asp174 ) exhibited diminished D1 receptor cell-surface expression

that was comparable to that observed with tunicamycin [14]. The findings from

the receptor mutants indicate that mutation of both glycosylation sites, and hence a

complete loss of glycosylation, may be required to attenuate D1 receptor trafficking.

However, given that the sole mutation of the Asp174 residue was not performed, an

important role for this individual site in the cell-surface transport of the D1 receptor

cannot be excluded.

Unlike the D1 receptor, the inhibitor tunicamycin has been shown to completely

abolish the cell-surface localization of the D5 receptor in cells [34]. Mutation of

the three individual N-linked glycosylation sites revealed that it was the Asp7

residue that was critical for the cell-surface transport of this receptor as its mutation resulted in an almost complete abolishment of plasma membrane D5 receptor

expression [34].

The elimination of N-linked glycosylation by mutation or tunicamycin resulted

in a reduction of both total cellular or plasma membrane expression of D2 receptors [14, 35]. Additionally, post-ER glycosylation has also been implicated in the

discrete trafficking of the D2 receptor isoforms, D2short (D2S ) and D2Long (D2L ).

Using pulse-chase procedures it was demonstrated that, under certain conditions,

D2S was rapidly processed from a newly synthesized protein to a partially, and then

fully glycosylated mature state. In contrast, a significant amount (approximately

20%) of the D2L isoform was only partially glycosylated and remained intracellularly sequestered [36]. In concordance with these findings, it has recently been

demonstrated that, in the absence of agonist, D2S was predominantly localized to

the plasma membrane, whereas the D2L isoform was found both at the cell surface

and intracellularly [37]. Similarly, it has been demonstrated that D2L was retained

further upstream and more strongly than D2S in early compartments of the secretory pathway [35]. The D2S and D2L isoforms differ from one another by a 29

amino acid insertion found in the third intracellular loop of D2L . Therefore, it is

plausible that the differential trafficking rates and targeting of these receptor isoforms are associated with this sequence of amino acids. Specifically, it is possible

that this sequence, in full or in part, may function as a retainment motif, whereby

interactions of the motif with an as yet unidentified accessory protein may serve to

retain the D2L isoform intracellularly. Given that D2S and D2L , respectively, function

pre- and postsynaptically [38], and work together to mediate dopamine transmission, understanding the mechanisms underlying the differential trafficking of these



8



Regulation of Dopamine Receptor Trafficking and Responsiveness



199



two isoforms may have significant implications in understanding the physiological

regulation of dopamine transmission in the brain.



8.2.2 Stabilization of Dopamine Receptors at the Cell Surface

It has been demonstrated that GPCRs are not static within the plasma membrane

but can move in the plane of the membrane by the passive process of lateral diffusion [39–41]. One of the key factors that govern the dynamics of lateral diffusion

of GPCRs is their association with other cellular proteins. The formation of these

protein complexes can serve to restrict the movements of the GPCRs, effectively

stabilizing the receptors in specific microdomains within the membrane, such as the

synapse for example. Numerous protein–protein associations of this type have been

reported between dopamine receptors and a variety of cellular proteins, examples of

which are discussed below.



8.2.2.1 The NMDA-D1 Receptor Trap

Movement by lateral diffusion has been reported for several GPCRs, including vasopressin V2 [39], serotonin 1A [40], and dopamine D1 receptors [41]. In cultured

neurons, approximately 65% of D1 receptors are mobile with the remaining receptors anchored within the membrane [41]. This ratio of mobile to anchored receptors

is not fixed, but fluctuating, and has recently been shown to be influenced by other

receptors within the plasma membrane. One such receptor is the NMDA receptor whose activation has been reported to recruit D1 receptors to the cell surface

from intracellular compartments [42] and, furthermore, restrict the lateral movements of D1 receptors within the membrane [41]. The modulation of D1 receptor

flow dynamics by NMDA activation at the cell surface stems in part from the ability

of the two receptors to physically interact. Particular regions of the D1 receptor carboxyl tail have been shown to bind to the NR-1 and NR-2 subunits of the NMDA

receptor [43]. It is the interface with the NR-1 subunit, however, that has been identified as the critical region for both NMDA-mediated cell-surface trafficking and

trapping of D1 receptors [41]. With regard to the functional mechanisms underlying D1 receptor trapping, the influx of calcium that results from NMDA receptor

activation does not appear to be involved. Rather it appears to be the result of an

allosteric transformation, induced through occupation of the NMDA binding site,

which facilitates the interaction between the D1 receptor and the NR-1 subunit [41].

It has been established that NMDA receptors are stabilized at the membrane via

anchorage to the postsynaptic density. Thus, the formation of NMDA–D1 receptor

complexes functions in localizing and stabilizing the D1 receptor at the synapse.

Presumably, increased synaptic localization would make these receptors more susceptible to activation, culminating in enhanced signal transduction and neuronal

responsiveness.



200



M.L. Perreault et al.



8.2.2.2 Role of Scaffolding Proteins in Dopamine Receptor

Cell-Surface Stability

Perhaps one of the most well-known functions of scaffolding proteins is to maintain the structural integrity of the cell membrane. However, it is becoming more

evident that scaffolding proteins may also serve additional important functions,

one of which involves the stabilization of GPCRs at the plasma membrane. With

regard to dopamine receptors, several scaffolding proteins have been identified

that serve in this capacity. Two examples are protein 4.1 N and filamin A (actinbinding protein 280), both of which have been shown to influence the expression

of D2 and D3 receptors through interactions at their third intracellular loops

[44–47].

The association of a truncated mutant of protein 4.1 N with D2 and D3 receptors

significantly reduced the expression of either receptor in cells, results that implicate

a positive role for protein 4.1 N in the cell-surface stabilization of these receptors [44]. Similarly, the actin-binding protein filamin A positively contributed to

D2 receptor plasma membrane expression [46, 47], and moreover, has been shown

to play a role in enhanced D2 receptor-mediated signaling [45]. It is possible that

the functional mechanism underlying this increase in signaling stems from filamin

A-assisted formation of D2 receptor clusters at specific locales on the cell surface.

Such clusters may function to increase the efficiency of receptor–effector coupling

by aggregating components of the signaling pathway. Filamin A also has been

reported to contribute to efficient signaling and sequestration of the D3 receptor

[48], although to date, a role for this protein in D3 receptor cluster formation has not

been reported.

In addition to increasing receptor stability, scaffolding proteins can also destabilize receptors at the plasma membrane. The cytoskeletal subunit neurofilament-M

(NF-M), for example, has been shown to negatively affect the cell-surface expression of the D1 receptor [49]. Specifically, coexpression of NF-M with the D1

receptor in cells resulted in a significant reduction in D1 receptor cell-surface number and ligand-mediated cyclic AMP (cAMP) accumulation. Interestingly, receptors

that remained at the cell surface exhibited a resistance to agonist-induced desensitization. Although the mechanism underlying this insensitivity was not identified, it

was postulated that formation of the NF-M/D1 complex might preclude associations

of the D1 receptor with kinases and/or arrestin, two events that are fundamental in

the initiation of receptor trafficking into the cell.



8.3 Desensitization

8.3.1 D1 -Like Receptors

Desensitization of D1 -like receptors (D1, D5 ) has been extensively studied over the

past several years and indicates that dopamine-induced attenuation of signaling by

these receptors occurs within minutes of exposure [50–56]. As with the majority of



8



Regulation of Dopamine Receptor Trafficking and Responsiveness



201



GPCRs, the predominant form of D1 receptor desensitization has been identified as

being mediated through GRKs. Truncated mutant constructs of the rat D1 receptor

have shown that multiple residues located downstream of Gly379 in the distal carboxyl terminus regulated dopamine-mediated phosphorylation and desensitization

of the D1 receptor, which was suggested to reflect the removal of potential GRK2

and/or GRK3 phosphorylation sites [50]. Carboxyl terminal sequences located

upstream of Gly379 (between Cys351 and Gly379 ) were shown to be important for

phosphorylation but not for desensitization [50]. Site-directed mutagenesis studies

of the human D1 receptor, on the other hand, have provided evidence to suggest

that GRK2 acts as a critical regulator of rapid agonist-induced receptor desensitization through phosphorylation of a single motif containing the residues Thr360

and Glu359 in the proximal segment of the carboxyl terminus [51]. Both of these

studies have used differential methodology which may play a role in the discrepant

results observed. Site-directed mutagenesis studies may be a more reliable method

for identifying the importance of specific residues since there is little change in the

intact structure of the receptor. Carboxyl terminal truncations, however, can alter

the structure of the receptor which permits access to previously sterically hindered

receptor domains, such as the third intracellular loop.

The third cytoplasmic loop has also been implicated in desensitization of the D1

receptor. It was previously demonstrated that the mutation of specific residues in

the third intracellular loop did not affect desensitization of the D1 receptor [51].

However, a subsequent report has demonstrated that these same residues were

involved in D1 receptor phosphorylation and desensitization [57]. A possible reason for this discrepancy may be the use of differential cell lines, where one study

used CHO cells and the other HEK 293 cells. It has been shown that the rate of

agonist-induced desensitization of the D1 receptor in CHO cells occurs more slowly

than in other cell types [58]. Thus, it has been postulated that D1 receptor phosphorylation may be GRK isoform dependent and these isoforms may be lacking in the

CHO cell line [57].

Given the evidence demonstrating the importance of the carboxyl terminus and

third intracellular loop, it has been proposed that D1 receptor phosphorylation takes

place in both the carboxyl terminus and the third intracellular loop in a sequential manner, where primary phosphorylation of the carboxyl terminus is permissive

for secondary third intracellular loop phosphorylation, which then allows for the

desensitization response [57].

In contrast to GRK2 phosphorylation, which requires receptor activation, GRK4

has been shown to regulate the constitutive phosphorylation and desensitization

of the D1 receptor [59] suggesting that specific GRK isoforms may serve discrete

functions in the regulation of dopamine receptor activity.

Other kinases, such as protein kinase A (PKA), may also play a role in homologous or agonist-specific forms of GPCR desensitization (reviewed [11]). Although

it has been demonstrated that the mutation of a potential D1 receptor PKA phosphorylation site reduced the rate of agonist-induced desensitization [60], and moreover,

that D1 receptor desensitization was blunted in cells deficient in PKA [58], it

has also been shown that the inhibition of PKA, either by substitution mutations



202



M.L. Perreault et al.



[51, 52] or pharmacologically [52], appeared to have no effect on D1 receptormediated increases in cAMP.

The D5 receptor exhibits very high sequence homology with the D1 receptor;

however, there are major differences in the intracellular loops and carboxyl termini.

Unlike the D1 receptor, the D5 receptor exhibits higher levels of constitutive activity

[61–63], a characteristic that has been shown to be regulated by the third cytoplasmic loop [64], as well as sequence-specific motifs within the carboxyl terminus

[61, 62]. Similarly, the D5 receptor also exhibits higher affinity for agonists, such

as dopamine, than the D1 receptor (reviewed [65]). A series of truncation/deletion

mutants of the D5 receptor identified the region encoded by amino acids 438–448

and particularly Gln439 as necessary and sufficient for full expression of higher

agonist affinities and constitutive activity relative to the D1 receptor. The last 40

amino acids of the D5 receptor, on the other hand, were shown unnecessary for the

observed distinguishing pharmacological and functional characteristics [61]. Given

that the carboxyl terminus has been implicated in a variety of GPCR regulatory

events (reviewed [66]) elucidation of further motifs or residues within the carboxyl

terminus of the D5 receptor is needed to help delineate the specific mechanisms

underlying D5 receptor trafficking.



8.3.2 D2 -Like Receptors

Early studies examining the functional desensitization of D2 -like receptors (D2 , D3 ,

D4 ) have generated variable results but, in general, indicate that they desensitize

much more slowly than D1 -like receptors and require prolonged agonist treatment

[53, 67]. Similar to the D1 receptor, the mechanisms underlying D2 -like receptor desensitization appear to involve GRKs, although their role in D3 receptor

desensitization as yet remains uncertain. Only by overexpression of GRK2, GRK5

[68], or GRK3 [69] was there increased phosphorylation of the human D2 receptor and receptor internalization, indicating the sensitivity of the D2 receptor as a

substrate for GRK phosphorylation is lower than the D1 receptor. These kinases,

however, did not appear to influence phosphorylation and desensitization of the D3

receptor [48, 69]. Indeed, only when D3 chimeras were generated containing the

second and third cytoplasmic loops of the D2 receptor, was GRK-mediated phosphorylation evident, possibly revealing the importance of these receptor domains

in GRK functioning [69]. It has also been reported, however, that GRK2 and

GRK3 levels may regulate the stability of the D3 receptor interaction with filamin

A [70], a scaffolding protein that has been shown to be involved in the stability

of D3 receptor expression at the plasma membrane [48]. In addition, gene deletion of GRK6 was shown to lead to enhanced coupling of D2 -like receptors to

their respective G proteins in vivo, an effect that was associated with increased

susceptibility to the locomotor-activating effects of psychostimulants [71], and suggests that GRK6 plays a role in regulating the responsiveness of the D2 and/or D3

receptors.



8



Regulation of Dopamine Receptor Trafficking and Responsiveness



203



The second messenger kinase, protein kinase C (PKC), has also been suggested to regulate the D2 and D3 receptors in a heterologous manner since PKC

activation was shown to attenuate the ability of both these receptors to inhibit

cAMP accumulation [48, 72]. PKC phosphorylation of the D2 receptor was demonstrated to take place on two internal domains within the third intracellular loop,

but only one residue, Ser355 , was shown to be involved in the PKC-induced

desensitization response [72]. Site-directed mutagenesis of all the possible phosphorylation sites within the intracellular loops of the D3 receptor identified Ser229

and Ser257 as the critical amino acids responsible for PKC-induced phosphorylation, desensitization, and internalization [48]. Additionally, PKC activation was

shown to induce specific effects on each D2 receptor isoform (D2L and D2S ) with

regard to receptor-stimulated calcium mobilization [73]. It has been reported that

although PKC is able to effectively desensitize D2S -induced increases in intracellular calcium, the D2L isoform is insensitive to PKC-induced desensitization

of calcium signaling due to the presence of a pseudosubstrate domain. A pseudosubstrate domain is a site that resembles a substrate domain except that the

serine phosphorylation site is replaced by alanine or other residues and therefore

may permit association with the kinase without resulting in functional phosphorylation [73]. This difference in substrate sensitivity of D2S and D2L appeared to

be the result of intramolecular competition between different substrate domains

on the D2L receptor for PKC recognition and a pseudosubstrate domain, which

is not found in the D2S receptor. Given the importance of the D2 receptor

in numerous physiological processes, the presence of pseudosubstrate domains

may potentially have significant implications for the regulation of the receptor

by PKC.



8.3.3 The D1 –D2 Heteromer

Although D1 and D2 receptors are biochemically and functionally distinct, some

physiological functions require the coactivation of both receptors [74, 75]. At a

mechanistic level this has been difficult to reconcile since coactivation of the D1 and

D2 receptors can result in both opposing and synergistic physiological responses.

The recent discovery, however, of a common functional output generated by the

concurrent activation of D1 and D2 receptors within the same cells resulting in

activation of a novel Gq/11-linked phospholipase C-dependent calcium signal [76]

has provided a possible biochemical mechanism by which the D1 and D2 receptors

work in concert to mediate these molecular and behavioral functions. Additionally,

in cultured cells coexpressing both receptors, the existence of D1 –D2 heteromers

was established by fluorescence resonance energy transfer [77], cotrafficking studies [77], and visualization of D1 –D2 heteromers in live cells [78]. Accordingly, a

heteromeric D1 –D2 signaling complex that could rapidly activate the Gq/11 protein

and result in intracellular calcium release was demonstrated to exist in the adult

rodent striatum [76, 77, 79].



204



M.L. Perreault et al.



Although little is known regarding the regulation of D1 –D2 heteromer responsiveness, it has been shown that desensitization of the agonist-induced calcium

signal occurs within minutes of agonist exposure and is initiated by agonist

occupancy of either receptor subtype, even though the signal is generated only

by occupancy of both receptors [80]. Additionally, the attenuation of receptor

internalization did not result in a concomitant decrease in the extent of signal desensitization, suggesting desensitization of the signal occurred prior to recruitment of

the complex into vesicles by endocytic machinery. Although GRK5 or GRK6 or any

of the second messenger kinases did not play a role in the desensitization, GRK2

and GRK3 appeared to have a role in the extent of desensitization. Inhibition of

GRK-mediated phosphorylation, however, did not inhibit this desensitization [80],

suggesting that, in addition to phosphorylating receptors, GRKs may also mediate signal desensitization by phosphorylation-independent mechanisms. It has been

suggested that GRK2 and GRK3 may sequester Gq/11 proteins, which interact with

the RGS domain on these GRKs [81]. Thus, this may provide a mechanism by which

GRK2 and GRK3 contribute to desensitization of the calcium signal mediated by

the D1 –D2 receptor heteromer [80]. It is of note, however, that heteromeric D1 and

D2 receptors exhibit conformations that permitted cross-phosphorylation of the D2

receptor by D1 receptor activation [77], a finding that implicates a discrete mechanism by which the D1 receptor within the D1 –D2 complex may regulate heteromer

functioning.



8.4 Internalization

8.4.1 D1 -Like Receptors

The acute administration of dopamine agonists has been demonstrated to induce

robust internalization of the D1 receptor in both cultured cells and neurons [82, 83],

as well as in vivo [84]. While in the absence of agonist the D1 receptor remained

predominantly on the cell surface, the addition of dopamine induced rapid internalization of approximately 70% of the receptors, with a half-life of less than 5 min

[56, 85]. Although endocytosis of the D1 receptor has been consistently documented

in both heterologous expression systems and neuronal cultures, the underlying

mechanisms have shown to be more variable. While earlier studies have identified

a role for PKA-mediated internalization in cells endogenously expressing the D1

receptor [86], mutagenesis of the PKA sites of the human D1 receptor [51], the

rat D1 receptor [60], and the non-human primate D1 receptor [52] did not affect

agonist-induced internalization.

Consistent with the role of GRKs in D1 receptor desensitization, this group of

kinases appears to play an essential role in D1 receptor internalization, although

the residues identified as being important for desensitization are not the same as for

internalization. Receptor mutagenesis has revealed that specific residues in the distal

portion of the carboxyl terminus (Thr446 , Thr439 , and Ser431 ) are involved in GRK2mediated internalization of the human D1 receptor [51]. However, rat D1 receptor



8



Regulation of Dopamine Receptor Trafficking and Responsiveness



205



mutants with carboxyl terminal truncations implied that sequences located between

Cys351 and Gly379 are pivotal to receptor internalization [50]. Although there appear

to be discrepancies regarding the relative importances of specific residues in D1

receptor internalization, the carboxyl terminus seems to be essential in this stage of

the endocytic trafficking pathway. It has also been postulated, however, that GRKmediated D1 receptor phosphorylation on the third intracellular loop may be of

relevance in promoting receptor interactions with arrestins [57], adaptor proteins

that have been shown to be essential for the internalization of a number of GPCRs

including the D1 receptor (reviewed [87]). Specifically, it has been suggested that

the phosphorylation of residues within the carboxyl terminus and third intracellular loop dissociates the two domains, allowing for arrestin to bind to the activated

third loop [57]. Activation of the D1 receptor leads to translocation of both arrestin2

and arrestin3 to the cell membrane, with arrestin3 being the more predominant

translocated subtype. Following arrestin membrane localization, the D1 receptor is

internalized and arrestin subsequently dissociates from the receptor at or near the

membrane [57, 88, 89]. Similarly, colocalization between endogenous D1 receptors

and arrestins in rat neostriatal neuronal cultures demonstrated that the D1 receptor

preferentially interacts with arrestin3 [90].

In addition to arrestins, studies assessing the internalization pathway of D1 receptor membrane trafficking have demonstrated the involvement of numerous other

proteins, including the scaffolding proteins PSD-95, clathrin, and caveolin-1, and

the GTPase dynamin [85, 91, 92]. In cultured cells, the coexpression of PSD-95 with

the D1 receptor resulted in a robust internalization of the receptor in the absence of

agonist. Additionally, the abolishment of PSD-95 in mice accentuated D1 receptormediated behavioral responses, suggesting that PSD-95 may also serve an inhibitory

role in the regulation of D1 receptor signaling in vivo [92]. Evidence suggests that

facilitation of D1 receptor internalization by PSD-95 is mediated through interactions with the carboxyl terminus of the D1 receptor and furthermore is dependent

on the presence of dynamin [92]. As dynamin has been previously shown to be

involved in dopamine-induced clathrin-mediated endocytosis of the D1 receptor

[56, 85], these findings implicate the clathrin-mediated endocytic pathway in the

internalization of the D1 receptor.

In addition to clathrin-mediated internalization, it has been shown in cultured

cells that the D1 receptor can be localized to low-density caveolin-enriched membrane domains and can associate with caveolin-1 in rat brain through a specific

binding motif found in transmembrane domain 7 [91]. Agonist stimulation of

the D1 receptor caused its translocation into caveolin-1-enriched membrane fractions, which was determined to be the result of D1 receptor endocytosis through

caveolae. However, unlike the relatively rapid clathrin-dependent mechanism of

internalization in which approximately 70% of activated receptors were internalized within 5 min [85], caveolin-dependent D1 receptor endocytosis appeared to

be kinetically slower, reaching approximately 55% internalization within 45 min of

agonist stimulation [91]. These findings suggest that both clathrin- and caveolinmediated processes may play functionally distinct roles in regulating D1 receptor

responsiveness in vivo. It would be of clinical relevance to determine whether



206



M.L. Perreault et al.



the relative contribution of each of these pathways differs in specific regions of

the brain.



8.4.2 D2 -Like Receptors

The endocytosis of the D2 receptor is a highly complex process that has been shown

to be both isoform and cell specific, as well as to exhibit both dynamin-dependent

and independent mechanisms [1, 57, 69, 85, 93].

Internalization of the D2 receptor requires increased levels of GRKs in heterologous cells and appears to be a relatively slow process taking approximately 2 h to

plateau [68, 93, 94]. Whereas little or no internalization was observed in the absence

of exogenous GRKs or in the presence of the dominant negative GRK2 (DN-GRK2),

coexpression of GRK2, GRK5 [68], or GRK3 [69] caused significant D2 receptor

internalization.

Similar to the D1 receptor, internalization of the D2 receptor involves GRKdependent receptor phosphorylation, followed by the translocation of arrestin2 and

arrestin3 to the cell membrane [69, 95] which function to promote receptor internalization [57]. The endogenous dopamine D2 receptor in neurons, however, has

been shown to preferentially interact with arrestin2 [95]. The D2 receptor isoforms

also showed differential regulatory mechanisms for internalization. For example,

although both isoforms displayed a similar level of phosphorylation and arrestin

translocation, the actual internalization of the two isoforms was differentially regulated by GRKs and arrestins, where the internalization of the D2S receptor was

preferentially enhanced by GRK2 or GRK3, but the D2L receptor was preferentially

enhanced by arrestin3 [96]. As discussed previously, given that the two receptor

isoforms differ by a 29 amino acid insertion in the third intracellular loop of the

D2L receptor, it is plausible that this region may play a role in isoform-specific

trafficking.

In contrast to the D1 receptor, D2 receptor internalization appears to be mediated

by specific dynamin isoforms, suggesting specificity between dynamin isoforms and

dopamine receptor subtypes. It has been reported that the internalization of the D2S

receptor is dynamin dependent, implicating the clathrin-coated endocytic pathway

in the internalization of this receptor [57, 69, 93]. There are conflicting reports, however, as to the importance of dynamin-mediated mechanisms in the internalization

of the D2L receptor. While it has been suggested that the D2L receptor internalizes in a dynamin-independent manner [57, 85], these studies assessed only the role

of the dynamin1 isoform, whereas the dynamin2 isoform has been more recently

implicated. In cultured cells and primary striatal neurons dynamin2 was shown to

localize to sites of D2 receptor internalization and associate with the D2 receptor

in the rat brain [1]. Furthermore, when high-resolution immunoelectron microscopy

was used to study internalization patterns of the D2 receptor in the primate prefrontal

cortex, the D2 receptor was demonstrated to undergo clathrin-mediated endocytosis

via clathrin-coated pits and clathrin-coated vesicles [97].

In contrast to the D2 receptor, the D3 receptor demonstrated little internalization in response to dopamine stimulation and only in the presence of overexpressed



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

2 Biosynthesis, Export, and Cell-Surface Stabilization

Tải bản đầy đủ ngay(0 tr)

×