Endocytosis of G Protein-Coupled Receptors Is Regulated by Clathrin Light Chain Phosphorylation

Current Biology 22, 1361–1370, August 7, 2012 ª2012 Elsevier Ltd All rights reserved

DOI 10.1016/j.cub.2012.05.034

Article Endocytosis of G Protein-Coupled Receptors Is Regulated by Clathrin Light Chain Phosphorylation Filipe Ferreira,1,3 Matthew Foley,1,3 Alex Cooke,2 Margaret Cunningham,2 Gemma Smith,1 Robert Woolley,1 Graeme Henderson,2 Eamonn Kelly,2 Stuart Mundell,2 and Elizabeth Smythe1,* 1Department of Biomedical Science, Centre for Membrane Interactions and Dynamics, University of Sheffield, Sheffield S10 2TN, UK 2School of Physiology and Pharmacology, University of Bristol, Bristol BS8 1TD, UK

Summary Background: Signaling by transmembrane receptors such as G protein-coupled receptors (GPCRs) occurs at the cell surface and throughout the endocytic pathway, and signaling from the cell surface may differ in magnitude and downstream output from intracellular signaling. As a result, the rate at which signaling molecules traverse the endocytic pathway makes a significant contribution to downstream output. Modulation of the core endocytic machinery facilitates differential uptake of individual cargoes. Clathrin-coated pits are a major entry portal where assembled clathrin forms a lattice around invaginating buds that have captured endocytic cargo. Clathrin assembles into triskelia composed of three clathrin heavy chains and associated clathrin light chains (CLCs). Despite the identification of clathrin-coated pits at the cell surface over 30 years ago, the functions of CLCs in endocytosis have been elusive. Results: In this work, we identify a novel role for CLCs in the regulated endocytosis of specific cargoes. Small interfering RNA-mediated knockdown of either CLCa or CLCb inhibits the uptake of GPCRs. Moreover, we demonstrate that phosphorylation of Ser204 in CLCb is required for efficient endocytosis of a subset of GPCRs and identify G protein-coupled receptor kinase 2 (GRK2) as a kinase that can phosphorylate CLCb on Ser204. Overexpression of CLCbS204A specifically inhibits the endocytosis of those GPCRs whose endocytosis is GRK2-dependent. Conclusions: Together, these results indicate that CLCb phosphorylation acts as a discriminator for the endocytosis of specific GPCRs. Introduction Clathrin-dependent endocytosis (internalization) is required for the trafficking and signaling of a range of receptors including the vast majority of the >800 G protein-coupled receptors (GPCRs) in the human genome [1]. Clathrin assembles from the cytoplasm at specialized regions of the cell surface called clathrin-coated pits (CCPs). The CCPs invaginate and pinch off to form clathrin-coated vesicles (CCVs). Following CCV uncoating, the endocytic vesicle is targeted to early endosomes where the cargo proteins are sorted to

3These authors contributed equally to the work *Correspondence: [email protected]

a variety of intracellular destinations [2]. It was originally thought that endocytosis acted to constitutively attenuate receptor signaling through removal of receptors from the cell surface, but an emerging paradigm is that subcellular localization and dwell time of signaling molecules on the endocytic pathway can be key to determining signaling output [3, 4]. At distinct locations on the endocytic pathway, different complements of signaling components (adaptors, kinases, phosphatases) will be recruited, thus allowing a single signal to give rise to multiple downstream outputs and allowing sophisticated modulation of the signaling process. For GPCRs, intracellular sorting is key to the balance between the clinically relevant processes of desensitization and resensitization [1]. Clathrin is involved not only in the endocytosis of cellsurface receptors, but it also functions in the delivery of newly synthesized lysosomal proteins from the biosynthetic pathway to late endosomes and recycling of cargo from endosomes back to the cell surface. In addition, flat clathrin lattices facilitate the sorting of ubiquitinated cargo to multivesicular bodies [2, 5, 6]. Clathrin exists in solution as a triskelion composed of three clathrin heavy chains (CHC, MW: 190 kDa) and three clathrin light chains (CLC, MW: w23–26 kDa). Budding yeast have a single clathrin light chain (Clc1p) and Clc1p null cells exhibit growth defects as well as reduced rates of receptormediated endocytosis [7]. All metazoans have two light chain genes, CLCA and CLCB, which have been highly conserved throughout evolution with approximately 65% homology between CLCA and CLCB and greater than 90% homology between CLCs from different species [8]. To date, the function(s) of the CLCs in metazoans has remained more elusive. They have been implicated in the trimerization of CHCs to form triskelia [9], and recent crystallographic data have indicated that the CLCs inhibit clathrin assembly when bound by holding the ‘‘knee’’ of CHC in a straight conformation, which is refractory to assembly [10]. However small interfering RNA (siRNA)-mediated knockdown of either or both CLCs had no discernible effect on CCP assembly or receptor-mediated endocytosis in mammalian cells [11–13], although recent studies showed that depletion of the CLCs affects trafficking of the mannose-6 phosphate receptor from the trans-Golgi network (TGN) to late endosomes [13]. In vitro clathrin disassembly occurs by the combined action of hsc70, which is recruited by a DnaJ domain containing protein, auxilin (neuronal cells), or GAK (cyclin G associated kinase) (nonneuronal cells) [14]. Auxilin/GAK binds to CHC [15], and there is a burst of auxilin recruitment just prior to CCV scission [16, 17], consistent with its role in uncoating. Initial in vitro studies had identified a role for CLCs in the recruitment of hsc70 to CCVs [18]. However, subsequent analysis of the role of auxilin indicated that CLCs were dispensable for auxilin-mediated uncoating of clathrin from cages by hsc70, at least, in vitro [19]. Following agonist stimulation, GPCRs can endocytose through CCPs. In most cases, receptors are phosphorylated by G protein-coupled receptor kinases (GRKs) leading to enhanced association with arrestins, with concomitant attenuation of G protein-dependent signaling [20]. Following binding

A 04 D S2

llA

04 S2

Sa

llA S2 04 S2 A 04 wt D

Sa

S2

Sa

wt

Co

nt

llA

ro

l

A

04 A S2 04 D wt

Current Biology Vol 22 No 15 1362

190kD

IB: α-CHC

26kD

IB: α-FLAG Lysates

α-FLAG

α-N/S

S2

Sa

wt

Co

nt

llA

ro

l

B

04 A S2 04 D Co nt ro wt l Sa llA S2 04 A S2 04 Co D nt ro l wt Sa llA

IP

IB: α-CHC

190kD

IB: α-CLCs

26kD 23.4kD 23kD Lysates

α-CHC

α-N/S

IP

C

Clathrin

CLCb

CLCb

CLCb

CLCb

FLAG-CLCb

Merge

wt

SallA

S204A

S204D

Figure 1. Phosphorylation Mutants of CLCb Associate with CHC (A and B) HEK293T cells were transfected with FLAG-tagged CLCbwt, CLCbSallA, CLCbS204A, or CLCbS204D, and lysates were prepared and subjected to an immunoprecipitation protocol using nonspecific mouse IgG (a-N/S) and either mouse anti-FLAG antibodies (A) or anti-CHC (X22) antibodies (B). The immunoprecipitates were probed for the presence of CHC using CHC5.9 antibodies, FLAG using rabbit anti-FLAG antibodies, or CLCs using anti-CLCa and anti-CLCb antibodies. Lysates represent 10% of the input material. Arrows indicate the positions of CHC (190kD), CLCa (23.4kD), CLCb (23kD), and FLAG-CLCb (26kD). (C) Immunofluorescence images of HEK293T cells transfected with CLCbwt, CLCbSallA, CLCbS204A, or CLCbS204D. CHC was detected with X22 antibody (green) and overexpressed CLCb (WT and mutants) was detected with rabbit anti-FLAG antibody (red). Scale bar represents 5 mm. See also Figures S1 and S2.

to GPCRs, arrestins can interact with CHC and the AP2 adaptor complex, thus facilitating recruitment of GPCRs to CCPs and subsequent endocytosis [21]. In some cases, GPCRs can traffic into CCPs in an arrestin-independent manner [22, 23]. Following delivery to endosomes, GPCRs may be rapidly recycled, resulting in resensitization, or targeted to lysosomes for degradation [22]. Furthermore, GPCRs can continue to signal in a G protein-independent manner following delivery to endosomes [3].

There is however considerable evidence to support a more complex regulation of GPCR signaling and trafficking, which gives exquisite spatiotemporal regulation [3, 20]. Different agonists can result in distinct trafficking outcomes, e.g., the endocytosis of the m opioid receptor (MOPr), which displays ligand-dependent trafficking into CCPs [23]. There is also evidence that GPCRs can be sorted into different classes of CCPs [24] that regulate GPCR uptake by different mechanisms. For example, the purinergic GPCRs P2Y1 and P2Y12 are both endocytosed via clathrin-dependent pathways, but whereas P2Y12 requires GRK2 and arrestins, P2Y1 trafficking is independent of these components but requires protein kinase C (PKC) [25]. Not all CCPs are uniform in their ability to recruit GPCR cargo [24, 25] and the molecular mechanisms underlying this phenomenon are largely unknown. In this study, we report a novel role for CLCs in the uptake of GPCRs. We demonstrate that CLCb phosphorylation is important for endocytosis where it contributes to efficient uncoating of CCVs. Furthermore we show that phosphorylation of CLCb, most likely by GRK2, may play a critical role in the segregation of GPCRs into distinct subpopulations of CCPs. In addition, we highlight a potentially important novel role for GRK2 in regulating GPCR trafficking downstream of receptor phosphorylation. Results Phosphorylation of CLCb Regulates Endocytosis Previous studies had shown that siRNA-mediated knockdown of CLCs did not affect transferrin uptake [11–13]. In order to understand the functions of CLCs, we adopted an alternative approach of asking whether CLC phosphorylation is important in the regulation of endocytosis. CLCb is phosphorylated both in vivo and in vitro, predominantly on serine residues [26, 27]. There are 19 serine residues in bovine CLCb (see Figures S1A and S1B available online). The amino acids referred to in this work refer to the positions in neuronal bovine CLCb. As a first step to investigate the role of CLCb phosphorylation, we generated a FLAG-tagged mutant form of bovine CLCb where all 19 serine residues were converted to alanines, CLCbSallA. Transfection of HEK293T cells with CLCbwt and CLCbSallA resulted in significant levels of overexpression compared to endogenous CLCb (Figure 1B). We verified that immunoprecipitation of both wild-type (WT) and mutant protein with anti-FLAG antibodies resulted in the coprecipitation of similar amounts of CHC. Under conditions where <10% of the FLAG-tagged CLCs (CLCbwt: 4%; CLCbSallA: 8%; CLCbS204A: 7%; and CLCS204D: 5%) was immunoprecipitated, a significant amount of CHC (CLCbwt: 11%; CLCbSallA: 22%; CLCbS204A: 18%; and CLCS204D: 18%) was coimmunoprecipitated indicating that the recombinant CLCs could associate with CHC in intact cells (Figure 1A). Reciprocally, immunoprecipitated CHC showed an approximately 60:40 ratio of FLAG-tagged CLCs to endogenous CLCa and CLCb (Figure 1B: CLCbwt: 68%; CLCbSallA: 59%; CLCbS204A: 61%; and CLCS204D: 65%). To confirm that CLCbSallA could assemble into CCPs, we examined the colocalization of overexpressed CLCs with antibodies against CHC (X22) (Figure 1C) and found good colocalization of FLAG-tagged CLCs with CHC (CLCbwt: 91.2%; CLCbSallA: 88.7%; CLCbS204A: 92.2%; CLCbS204D: 90.1%). There was similarly good colocalization with TfR and AP2 (Figure S2A). Furthermore, recombinant CLCs could assemble into reconstituted clathrin cages in vitro (Figure S3).

B

70 60 50 40

CLCb wt CLCb SallA

30 20 10 0

0

5

10

15

20

25

30

% cell-associated transferrin

A

% Transferrin internalised

Clathrin Light Chains Regulate GPCR Endocytosis 1363

120 100 80

CLCb wt CLCb SallA

60 40 20 0

0

5

10

% Transferrin internalised

Time (min)

60

20

25

30

C

*

50

15

Time (min)

CLC

wt

CLC

SallA

CLC

S204A

40

*

30

CLCb wt CLCb S204A

20 10 0

0

5

10

15

20

25

CLC wt +CIP

30

% Transferrin internalised

Time (min)

50

*

40 30 20

CLCb wt CLCb S204D

10 0

0

5

10

15

20

25

30

Time (min) Figure 2. Phosphorylation of CLCb on Ser204 Modulates Endocytosis (A) HEK293T cells were transfected with CLCbwt, CLCbSallA, CLCbS204A, or CLCbS204D and the rate of transferrin endocytosis was measured as described in Experimental Procedures. Results are expressed as the percentage transferrin internalized as a function of time, and each point represents the mean 6 SEM of at least three independent experiments (*p < 0.05). (B) HEK293T cells were transfected with CLCbwt or CLCbSallA, and the rate of transferrin recycling was measured as described in Experimental Procedures. Data are expressed as the mean 6 SD of two separate experiments each performed in duplicate. (C) Lysates were prepared from HEK293T cells transfected with FLAG-tagged CLCbwt, CLCbSallA, or CLCbS204A and subjected to immunoprecipitation with anti-FLAG antibodies. The immunoprecipitates were subjected to 2D gel electrophoresis and western blotting with anti-FLAG antibodies. A sample of the immunoprecipitate from lysates expressing CLCbwt was incubated with calf intestinal alkaline phosphatase (CIP) as described in Experimental Procedures. Arrows indicate different phosphorylated forms of CLCb that disappear on treatment with CIP. Asterisk indicates a spot that is absent in CLCbS204A. See also Figure S2.

We next asked whether overexpression of CLCbSallA affected endocytosis by measuring the rate of transferrin internalization in cells transfected with either CLCbwt or CLCbSallA. Whereas cells transfected with CLCbwt showed similar kinetics of endocytosis compared to control cells (Figure S2B), overexpression of CLCbSallA caused a dominant-negative effect on endocytosis with a reduction (w30%) in the rate of transferrin internalization (Figure 2A, top panel). In contrast, recycling from early endosomes was unaffected by CLCbSallA overexpression (Figure 2B), arguing against a global disruption of the endocytic pathway. In order to determine which specific residue(s) was important for the dominant-negative effect, we generated a series

of constructs where groups of serines were mutated to alanine. By examining a range of constructs containing combinations of mutations and an iterative process (Figure 2A; Figures S3B and S3C; data not shown), we were able to establish that Ser204 is necessary and sufficient to mediate the dominant-negative effect (Figure 2A, second panel). As shown in Figure 1C, CLCbS204A also coimmunoprecipitates with CHC and colocalizes with CCP components to the same extent as CLCbSallA. We also generated a phosphomimetic mutant, CLCbS204D, and confirmed that it associated with clathrincoated structures by immunofluorescence and immunoprecipitation (Figure 1). When tested in the transferrin internalization assay, we observed a mild inhibition of transferrin

Current Biology Vol 22 No 15 1364

Clathrin

A

CLCb

CLCb

CLCb

Transferrin

Merge

Figure 3. Overexpression of CLCbS204A Causes a Delay in CCV Uncoating (A) HEK293T cells, transfected with CLCbwt or CLCbS204A or CLCbS204D were incubated with Tx-Red Tfn for 5 min at 37 C. Surface Tx-Red Tfn was acid stripped and the cells were fixed and CHC was visualized using the mAb X22 (green). Arrows indicate an overlap between X22 and Tx-Red Tfn. Scale bar represents 2 mm. (B) Quantification of the degree of overlap of Tx-Red Tfn and X22 in cells overexpressing CLCbwt or CLCbS204A or CLCbS204D. The degree of colocalization in cells overexpressing CLCbwt has been arbitrarily set to 1. Results are the mean 6 SEM (CLCbwt: 73 cells; CLCbS204A: 93 cells; CLCbS204D: 100 cells). Values are significantly different at **p < 0.01 or *p < 0.05. See also Figure S3.

wt

S204D

intestinal alkaline phosphatase. None of these spots were present in the immunoprecipitate where all of the serines were mutated to alanines (CLCbSallA), whereas the absence of one of the spots in CLCbS204A provides strong evidence that Ser204 is phosphorylated in cells.

S204A

Clathrin and transferrin overlap wt Arbitrary units normalised to CLCb

Phosphorylation of Ser204 Is Required for Efficient Uncoating in Intact Cells B We next addressed how CLCbS204A overexpression might inhibit transferrin endocytosis. To do this, we used an 1.8 assay in intact cells that measures clathrin uncoating [28]. Briefly, transfected 1.6 HEK293T cells are incubated for 5 min 1.4 with Texas-Red transferrin (Tx-Red Tfn). Cell surface Tx-Red Tfn is acid1.2 stripped and then the degree of colocalization of clathrin with endocytosed 1 Tx-Red Tfn is measured to indicate its 0.8 association with endocytic vesicles. CCVs rapidly lose their coats following 0.6 scission, and thus the degree of colocalization of Tx-Red Tfn with clathrin is low 0.4 (<10%, Figure 3A). We thus arbitrarily 0.2 normalize this value to 1 in cells overexpressing CLCbwt. Under conditions 0 where uncoating is delayed, there is an wt S204D S204A increase in the degree of colocalization CLCb CLCb CLCb of clathrin with internalized Tx-Red Tfn [28]. Following transfection of HEK293T internalization (Figure 2A, third panel), suggesting that cycles cells with CLCbwt or CLCbS204A or CLCbS204D, the degree of of CLCb phosphorylation are important for at least some of colocalization of clathrin with Tx-Red Tfn was measured and was found to be significantly higher following expression of its functions in endocytosis. To provide evidence that Ser204 is phosphorylated in cells, CLCbS204A. Overexpression of CLCbS204D also showed a delay we immunoprecipitated CLCb from cells transfected with in clathrin uncoating, which was significantly different from FLAG-tagged CLCbwt, CLCbSallA, and CLCbS204A and sub- CLCbwt but less than the effect observed following overexjected the immunoprecipitates to two-dimensional (2D) gel pression of CLCbS204A (100% 6 7% for CLCbwt, 124% 6 6% electrophoresis followed by blotting with anti-FLAG anti- for CLCbS204D, and 160% 6 8.2% for CLCbS204A) (Figure 3B). bodies (Figure 2C). Lysates from cells transfected with This indicates that CLCbS204A overexpression specifically CLCbwt showed four spots of increasing pI, which corre- slows clathrin uncoating. Clathrin uncoating is mediated by the combined action sponded to different phosphorylated forms of CLCb because they were greatly reduced following treatment with calf of hsc70 and its cofactor auxilin/GAK. We observed that

**

*

Clathrin Light Chains Regulate GPCR Endocytosis 1365

binding of auxilin to reconstituted cages containing CLCbS204A was stronger than to cages containing CLCbS204D (Figures S3A and S3B). We used total internal reflection fluorescence (TIRF) microscopy to investigate the behavior of GFP-GAK at individual clathrin spots in HeLa cells overexpressing mCherryCLCbwt, mCherry-CLCbS204A, or mCherry-CLCbS204D (Figures S4D–S4F). We observed a ‘‘burst’’ of GFP-GAK recruitment just prior to the disappearance of clathrin spots as described [16, 17]. There was a statistically significant decrease in the rate of dissociation of GFP-GAK from clathrin spots in HeLa cells in the presence of overexpressed mCherry-CLCbS204A compared to overexpression of mCherry-CLCbS204D or CLCbwt (Figure S3F). Together, these data suggest modulation of auxilin and clathrin interactions as a potential mechanism by which CLCb phosphorylation might interfere with endocytosis. Ser204 Is an In Vitro Substrate for GRK2 The next question that we wished to address was the identity of the kinase(s) that phosphorylates Ser204. In silico analysis (http://ppsp.biocuckoo.org/index.php) suggested that the sequence surrounding Ser204 might be a consensus site for G protein-coupled receptor kinases (GRKs). There are seven GRK family members: GRKs 1 and 7 are predominantly involved in the visual system, GRKs 2 and 3 have a PtdIns (4,5) P2-binding PH domain that allows association with the plasma membrane, and GRKs 4, 5, and 6 use covalent modification by lipids to interact with the plasma membrane [22]. Because the sequence surrounding Ser204 is also found in CLCa (Figure S1B), we tested whether GST-fusion proteins including CLCa and CLCb could act as substrates for GRKs using recombinant GRK2 and GRK5. Whereas GRK5 did not significantly phosphorylate CLCbwt (data not shown), GRK2 was capable of phosphorylating CLCbwt and CLCawt (Figure 4A) and, strikingly, phosphorylation was reduced (w50%, n = 3) although not abolished in CLCbS204A, suggesting that GRK2 phosphorylates not only Ser204 but also another site(s) on CLCb in vitro (Figure 4B). By contrast, CK2, which is one of the major CCV kinases and phosphorylates CLCs [29], did not appear to phosphorylate Ser204 (Figure 4B). To obtain more direct evidence that GRK2 phosphorylates CLCb on Ser204 in vivo, we used siRNA to deplete GRK2 from HEK293T cells that had been transfected with FLAGCLCbwt and FLAG-CLCbS204A. FLAG-CLCbs were then immunoprecipitated and subjected to 2D analysis as described for Figure 2C (Figure 4C). Knockdown of GRK2 was efficient (86% and 88% for cells transfected with CLCbwt and CLCbS204A, respectively). FLAG-CLCbwt, which was immunoprecipitated from cells treated with siRNA targeting GRK2, showed a phosphorylation pattern that differed from control cells in lacking at least one spot and which was very similar to the pattern seen in the immunoprecipitate of FLAGCLCbS204A. siRNA knockdown of GRK2 did not alter the pattern of spots seen in the immunoprecipitate of FLAGCLCbS204A. Together, these data provide strong evidence that GRK2 phosphorylates CLCb in cells and that Ser204 is a major site phosphorylated by GRK2.

peptide agonist, DAMGO, in HEK293T cells stably expressing hemagglutinin (HA) epitope-tagged receptor, using an enzyme-linked immunosorbent assay (ELISA) [23]. DAMGOinduced endocytosis is known to be GRK2-dependent and at receptor-saturating concentration (10 mM), DAMGO induced up to 30% loss of cell surface MOPr over 30 min (Figure 4D; [23]). Overexpression of either CLCbSallA (data not shown) or CLCbS204A (Figure 4D) strongly inhibited the DAMGO-induced cell surface loss of MOPr. This receptor also undergoes a small amount of endocytosis in response to the clinically important drug morphine, and as with DAMGO, expression of CLCbS204A inhibited morphine-induced endocytosis (Figure 4D). Together, these results indicate that mutation of Ser204 alone in CLCb is sufficient to mediate the dominantnegative effect on MOPr trafficking. By contrast, there was no effect on DAMGO- or morphine-induced MOPr endocytosis following overexpression of CLCbS204D. We next investigated whether Ser204 phosphorylation was the single phosphorylation site required for GPCR endocytosis or whether there were other potential regulatory sites. We decided to examine the purinergic P2Y receptors because these receptors populate different subclasses of CCPs and have different requirements for endocytosis [25]. P2Y1 endocytosis is PKC-dependent, whereas P2Y12 requires GRK2 and arrestins. 1321N1 astrocytoma cells stably expressing HA-tagged P2Y1 or P2Y12 were therefore transfected with vector alone (pcDNA), CLCbSallA, or CLCbS204A. Following incubation with ADP, the loss of cell surface receptor was assessed by an ELISA assay. In the case of P2Y12, overexpression of both CLCbSallA and CLCbS204A strongly inhibited the ADP-induced loss of cell surface receptor suggesting that for this receptor, mutation of Ser204 is necessary and sufficient to mediate the dominant-negative effect (Figure 4E). The endocytosis of the GRK-independent P2Y1 receptor was enhanced by overexpression of CLCbwt. Whereas P2Y1 internalization was inhibited by overexpression of CLCbSallA, overexpression of CLCbS204A did not significantly affect agonist-dependent receptor endocytosis (Figure 4E). This result indicates that other phosphorylation sites on CLCb are important for the regulation of endocytosis of those GPCRs, such as P2Y1, that are GRK2-independent. The strong dominant-negative effect of overexpression of CLCbSallA and CLCbS204A (for P2Y12) on the internalization of P2Y receptors implies a crucial role for CLCs in the endocytosis of these receptors. To test this, we used siRNA to knock down CLCa and CLCb in 1321N1 cells and then measured the ADP-induced endocytosis of both the P2Y1 and P2Y12 receptors (Figure 5). Knockdown of CLCs was >50% efficient (Figure 5A) and resulted in a striking inhibition of endocytosis of both P2Y1 and P2Y12 receptors (Figure 5B). Previous studies have indicated that knockdown of CLCs has no significant effect on the endocytosis of transferrin [11–13]. Together with the relatively mild effect of overexpression of CLCbS204A on transferrin uptake (Figure 2A), these results reinforce the importance of CLCs in the uptake of specific cargoes. Discussion

GPCR Endocytosis Is Regulated by Differential Phosphorylation of CLCb To investigate the functional significance of CLCb phosphorylation by GRK2, we asked whether the endocytosis of GPCRs, which are major in vivo substrates for GRKs, was affected by overexpression of CLCb phosphorylation mutants. We first examined the endocytosis of MOPr in response to the

Through studies on CLC phosphorylation, we have uncovered a novel role for CLCs and specifically, CLCb phosphorylation, in the regulation of endocytosis. We demonstrate for the first time that siRNA-mediated knockdown of the CLCs inhibits the uptake of particular endocytic cargo. We show that phosphorylation of CLCs is important for their function and,

Current Biology Vol 22 No 15 1366

48.4kD 48kD

Gel

48.4kD 48kD

Autorad b

a

110

GRK2

B

CK2 Gel

48kD 48kD

Autorad

CLCb

C

% Surface receptor

GST-CLC

D DAMGO

wt

S204A

wt

S204A

80kD

GRK2

FLAG-CLCbwt + FLAG-CLCbS204A − GRK2 siRNA −

− + −

+ − +

CLCb

wt

CLCb

S204A

CLCb

S204A

CLC wt CLC S204A

80

CLC S204D

70 60 10 0

0

5

10

15

20

25

110

wt

Control

*

90

30

Morphine

− + +

CLCb

100

Time (min)

% Surface receptor

A

+GRK2 KD

+GRK2 KD

*

100

Control CLC wt

90

CLC S204A CLC S204D

80 70 60 10 0

0

5

10

15

20

25

30

Time (min)

E P2Y12

P2Y1

% Surface receptor

90

*

80 70 10 0

0

5

*

* *

10

*

15

20

25

Control CLC wt CLC

SallA

CLC S204A

30

% Surface receptor

100

100

*

90

Control CLC wt CLC SallA

80

CLC S204A

70 60 50 10 0

* 0

5

10

15

20

25

30

Time (min)

Time (min)

Figure 4. Ser204 Phosphorylation Selectively Affects GRK2-Dependent GPCR Trafficking (A) GRK2 phosphorylates CLCa and CLCb: Coomassie-stained gels (upper) or autoradiograms (lower) of in vitro kinase reactions using either GST-CLCa or GST-CLCb as substrate and GRK2 as kinase. (B) GRK2 but not CK2 phosphorylates CLCb on Ser204: Coomassie-stained gels (upper) or autoradiograms (lower) of in vitro kinase reactions using either GST-CLCbwt or GST-CLCbS204A as substrate and GRK2 or CK2 as kinase. (C) GRK2 is an in vivo kinase for CLCb: western blot of HEK293T cells treated with GRK2 siRNA or mock-treated (upper panel). Lysates were prepared from control HEK293T cells or those transfected with siRNA against GRK2 and which had also been transfected with either FLAG-tagged CLCbwt or CLCbS204A and subjected to immunoprecipitation with anti-FLAG antibody. The immunoprecipitates were subjected to 2D gel electrophoresis and western blotting with anti-FLAG antibodies. (D) Agonist-induced cell surface loss of MOPr is reduced by overexpression of CLCbS204A. HEK293T cells stably expressing HA-tagged-MOPr were untransfected (control) or transiently transfected with CLCbwt, CLCbS204A, or CLCbS204D. Cells were stimulated with morphine (30 mM) or DAMGO (10 mM) for 0–30 min, and changes in cell surface receptor expression determined by ELISA. Surface receptor present at time 0 was taken as 100%. Values are means 6 SEM of four independent experiments. Asterisk indicates values that are significantly different from control value at the same time point, as assessed by two-way ANOVA with Bonferroni posttest, p < 0.05. (E) P2Y1 and P2Y12 cell surface loss show differential requirements for CLCb phosphorylation: 1321N1 cells stably expressing HA-tagged-P2Y12 or P2Y1 purinoceptor were untransfected (control) or transiently transfected with CLCSallA, CLCS204A, or CLCwt. ADP-induced (10 mM; 30 min) loss of surface receptor was assessed by ELISA. Expression of CLCbSallA and CLCbS204A significantly reduced ADP-induced cell surface loss of P2Y12, but only CLCbSallA reduced loss of P2Y1. Surface receptor present at time 0 was taken as 100%. Values are means 6 SEM of three to seven independent experiments for each construct. Asterisk indicates values that are significantly different from control value at the same time point, as assessed by two-way ANOVA with Bonferroni posttest, p < 0.05.

specifically, that overexpression of CLCbS204A results in a dominant-negative effect on the rate of endocytosis of cargo such as GPCRs. We present evidence to support

phosphorylation of CLCb on Ser204 by GRK2 and importantly those cargoes that require GRK2 activity for endocytosis are most strongly affected by CLCbS204A overexpression.

Clathrin Light Chains Regulate GPCR Endocytosis 1367

A

P2Y1 P2Y12 P2Y1 P2Y12

P2Y1 P2Y12 P2Y1 P2Y12

23.4kD 50kD

N/S siRNA CLCa siRNA CLCb siRNA

+ − −

B

+ − −

− + −

CLCa

23kD

CLCb

α-tubulin

50kD

α-tubulin

− + −

+ − −

+ − −

− − +

− − +

P2Y1

*

% Surface receptor

100

*

*

90

CLCb siRNA CLCa siRNA

*

80

70 10 0 0

5

10

15

20

25

20

25

Time (min)

P2Y12

% Surface receptor

N/S siRNA

100

*

90

*

* *

80 70 60 10 0 0

5

10

15

Time (min)

Ser204 Phosphorylation of CLCb Regulates Clathrin-Mediated Endocytosis siRNA-mediated ablation of CLCs has no apparent effect on the endocytic uptake of housekeeping receptors such as transferrin receptor [11, 12]. We therefore pursued an alternative approach to understand the functions of CLCs and, in particular, to address the question as to whether they might be regulated by reversible phosphorylation. Overexpression in HEK293T cells of a mutant form of CLCb where all of the serines had been mutated to alanines caused a reduction in the rate of transferrin endocytosis. By generating a series of constructs where different serines were mutated to alanines either singly or in combination, we were able to identify a single amino acid substitution, Ser204 to alanine, as being necessary and sufficient to mediate the dominant-negative effect. Two-dimensional gel analysis provided strong evidence that Ser204 is phosphorylated in intact cells. This analysis indicated that other sites are also phosphorylated on CLCb, although they do not appear to be involved in clathrinmediated endocytosis of transferrin. These other sites are likely to be relevant for the endocytosis of a subset of GPCRs,

Figure 5. siRNA-Mediated Knockdown of CLCa or CLCb in 1321N1 Cells Reduces ADP-Induced Loss of Surface P2Y1 and P2Y12 Purinoceptor Cells were transfected with siRNA targetting CLCa or CLCb, as described in Experimental Procedures. (A) Western blot (top panels) indicated significant knockdown of CLCa and CLCb; in two experiments for each, percent knockdown as compared to cells treated with control siRNA was, for CLCa, 96% and 86% for P2Y1 cells, 83% and 78% for P2Y12 cells, and for CLCb 79% and 91% for P2Y1 cells, 82% and 55% for P2Y12 cells. (B) ADP-induced (10 mM; 30 min) loss of surface HA-tagged-P2Y1 or P2Y12 purinoceptor was assessed by ELISA. Surface receptor present at time 0 was taken as 100%. Values are means 6 SEM of three independent experiments for each siRNA. Asterisk indicates values that are significantly different from nonspecific siRNA (control) value at the same time point, as assessed by two-way ANOVA with Bonferroni posttest, p < 0.05.

30

particularly those whose regulation is GRK-independent (see below). They may also regulate other clathrin-mediated events, e.g., sorting by flat clathrin lattices on endosomes or delivery of newly synthesized lysosomal proteins from the TGN to the endocytic pathway, * because the ability of CLCbSallA to assoN/S siRNA ciate with clathrin-coated structures CLCb siRNA argues against their having a structural * CLCa siRNA role in clathrin assembly. Furthermore, overexpression of CLCbSallA did not result in a global disruption of membrane traffic because recycling of transferrin from early endosomes was unaffected. Although the nature of the 2D gel ex30 periment does not allow us to measure quantitative changes in the level of phosphorylated spots in immunoprecipitated samples from different sets of transfected cells, we cannot rule out the possibility that Ser204 phosphorylation is the first site in a hierarchy of phosphorylation sites, all of which may contribute to the modulation of endocytosis. Cycles of phosphorylation and dephosphorylation are known to regulate endocytosis [30–32]. Overexpression of the phosphomimetic CLCbS204D did not affect MOPr endocytosis or the dissociation of GFP-GAK from clathrin patches, suggesting that for these endocytic events, CLCb is likely to be phosphorylated. CLCbS204D overexpression caused a mild inhibition of transferrin internalization and clathrin uncoating, which suggests a role for CLCb dephosphorylation at other stages in the CCV cycle. CLCb Phosphorylation Acts as a Discriminator for the Endocytosis of a Subset of Cargoes The identification of GRK2 as a kinase important for the phosphorylation of Ser204 of CLCb in vitro and in vivo has provided insight into the physiological importance of this event. GRK2 acts on a large number of GPCRs to phosphorylate them and activate downstream arrestin-dependent

Current Biology Vol 22 No 15 1368

Figure 6. CLCb Phosphorylation Acts as a Discriminator for Different CCP Populations CLCb phosphorylation is important for endocytosis of a range of housekeeping receptors such as transferrin receptor and signaling cargo such as GPCRs. GPCRs such as P2Y1 and P2Y12 enter cells via different populations of CCPs. Overexpression of CLCbSallA has a dominant-negative effect on endocytosis of several GPCRs. Phosphorylation of Ser204 is most important for those cargoes that are dependent on GRK2 for their endocytosis, e.g., P2Y12 and MOPr. This supports the existence of CCP subpopulations, selective for particular cargo. Housekeeping receptors such as transferrin receptor may enter cells through either population. The portal of entry into cells and the dwell time at a particular location is likely to significantly impact on the signaling output of cargoes such as GPCRs.

signaling pathways. Whereas overexpression of CLCbSallA inhibited the DAMGO-induced endocytosis of MOPr as well as that of P2Y1 and P2Y12 receptor endocytosis, CLCbS204A overexpression was necessary and sufficient to mediate the dominant-negative effect on the endocytosis of those receptors that are dependent on GRK2, i.e., P2Y12 and MOPr internalization. We have shown that knockdown of GRK2 is sufficient to abolish phosphorylation of a site on CLCbwt that is absent from CLCbS204A, indicating that GRK2 is indeed an in vivo kinase for CLCb. Our data indicate that other phosphorylation sites in CLCb play roles in the uptake of GRK-independent cargo such as P2Y1. Transferrin receptor is endocytosed irrespective of the presence or absence of bound transferrin [33] and is generally regarded as a constitutive cargo which, together with the abundance of transferrin receptors on the cell surface of most cells, suggests that transferrin receptor would be

included in all CCPs in a stochastic fashion [34]. Although overexpression of CLCbS204A mirrors the overexpression of CLCbSallA in the inhibition of transferrin uptake, the extent to which its uptake is inhibited is relatively mild compared to the more dramatic effects on GPCR endocytosis, suggesting that in HEK293T cells, transferrin receptor enters pits populated by GRK2-dependent and -independent cargo (Figure 6). Furthermore, the contrast between the strong effects on GPCR uptake with the milder effects on transferrin internalization supports the notion that CLCs modulate cargo-specific uptake. These data are also consistent with the differential effects on transferrin and GPCR uptake following knockdown of CLCs. Whereas siRNA-mediated knockdown of CLCs does not affect transferrin internalization [11–13], strikingly, knockdown of either CLCa or CLCb is sufficient to inhibit the endocytosis of both P2Y1 and P2Y12 receptors suggesting that both CLCs are required for uptake, raising the interesting possibility that they may have distinct functions in cargo-specific uptake. Furthermore, overexpression of CLCbwt enhanced P2Y1 uptake suggesting that either or both CLCs may be limiting for uptake of particular cargo, consistent with the observation that in nonneuronal cells, CLCs are present in substoichiometric amounts compared to CHC [35]. Most of our experiments in this study focused on the role of CLCb. However, CLCa contains the same consensus site for GRK2 phosphorylation and can be phosphorylated as efficiently by GRK2 in vitro. We therefore think it highly likely that phosphorylation of CLCa on this site will perform a similar function to CLCb phosphorylation. This may however be celltype-specific because in MDBK cells, CLCb is phosphorylated to a greater extent than CLCa [27]. The identification of an endocytic protein, CLCb, as a substrate for GRK2 is a novel finding and suggests that

Clathrin Light Chains Regulate GPCR Endocytosis 1369

GRK2 integrates GPCR trafficking and signaling by phosphorylating core endocytic components as well as cargo. GRK5 did not phosphorylate Ser204, suggesting interesting differences in the ability of different GRKs to direct or regulate endocytic pathways. Phosphorylation of CLCb may enhance the rate of uncoating of those CCVs that are enriched in GPCRs such as P2Y12 and MOPr. Regulation of uncoating of CCV populations enriched in GPCRs could contribute to intracellular signaling by GPCRs, e.g., by facilitating their delivery to an endosomal compartment specialized for activation of a particular signaling output. Alternatively, CLCb phosphorylation may ‘‘mark’’ a vesicle population for a particular trafficking pathway in the cell, e.g., the rapid recycling pathway followed by MOPr and the P2Y12 receptors, which leads to the relatively rapid resensitization of these GPCRs [36, 37]. In summary, we have revealed a previously unknown function for CLCs in endocytosis. Moreover, we have shown that differential phosphorylation of CLCb is an important modulator of the endocytosis of particular cargoes such as GPCRs, which is likely to have profound effects on downstream signaling.

8.

9.

10.

11.

12.

13.

14.

Experimental Procedures

15.

Cell culture and endocytosis assays were performed as previously described [20, 28]. Details of transfections, 2D gels, mutagenesis, and siRNA constructs are available in the online Supplemental Information.

16.

Supplemental Information

17.

Supplemental Information includes three figures and Supplemental Experimental Procedures and can be found with this article online at doi:10.1016/j.cub.2012.05.034.

18.

Acknowledgments We thank Tony Jackson, Christien Merrifield, Lois Greene, and Ernst Ungewickell for the provision of reagents and Kathryn Ayscough, Carl Smythe, and members of the Smythe laboratory for comments on the manuscript. We thank Zeiss for the use of their TIRF microscope. This work was supported by MRC (program grant G0300452), a BBSRC studentship (M.F.), an FCT studentship cofunded by POPH/FSE (F.F.), and a British Pharmacological Society AJ Clark studentship (A.C.). The microscopy was performed in the University of Sheffield Light Microscopy Facility (Wellcome grant GR077544AIA).

19.

20.

21.

22.

23. Received: August 26, 2011 Revised: March 16, 2012 Accepted: May 16, 2012 Published online: June 14, 2012

24.

References

25.

1. Wolfe, B.L., and Trejo, J. (2007). Clathrin-dependent mechanisms of G protein-coupled receptor endocytosis. Traffic 8, 462–470. 2. Conner, S.D., and Schmid, S.L. (2003). Regulated portals of entry into the cell. Nature 422, 37–44. 3. Sorkin, A., and von Zastrow, M. (2009). Endocytosis and signalling: intertwining molecular networks. Nat. Rev. Mol. Cell Biol. 10, 609–622. 4. Scita, G., and Di Fiore, P.P. (2010). The endocytic matrix. Nature 463, 464–473. 5. Brodsky, F.M., Chen, C.Y., Knuehl, C., Towler, M.C., and Wakeham, D.E. (2001). Biological basket weaving: formation and function of clathrincoated vesicles. Annu. Rev. Cell Dev. Biol. 17, 517–568. 6. Raiborg, C., Bache, K.G., Gillooly, D.J., Madshus, I.H., Stang, E., and Stenmark, H. (2002). Hrs sorts ubiquitinated proteins into clathrincoated microdomains of early endosomes. Nat. Cell Biol. 4, 394–398. 7. Chu, D.S., Pishvaee, B., and Payne, G.S. (1999). A modulatory role for clathrin light-chain phosphorylation in Golgi membrane protein

26.

27. 28.

29. 30.

localization during vegetative growth and during the mating response of Saccharomyces cerevisiae. Mol Biol Cell 10, 713–726. Jackson, A.P., and Parham, P. (1988). Structure of human clathrin light chains. Conservation of light chain polymorphism in three mammalian species. J. Biol. Chem. 263, 16688–16695. Ybe, J.A., Greene, B., Liu, S.H., Pley, U., Parham, P., and Brodsky, F.M. (1998). Clathrin self-assembly is regulated by three light-chain residues controlling the formation of critical salt bridges. EMBO J. 17, 1297–1303. Wilbur, J.D., Hwang, P.K., Ybe, J.A., Lane, M., Sellers, B.D., Jacobson, M.P., Fletterick, R.J., and Brodsky, F.M. (2010). Conformation switching of clathrin light chain regulates clathrin lattice assembly. Dev. Cell 18, 841–848. Hinrichsen, L., Harborth, J., Andrees, L., Weber, K., and Ungewickell, E.J. (2003). Effect of clathrin heavy chain- and alpha-adaptin-specific small inhibitory RNAs on endocytic accessory proteins and receptor trafficking in HeLa cells. J. Biol. Chem. 278, 45160–45170. Huang, F., Khvorova, A., Marshall, W., and Sorkin, A. (2004). Analysis of clathrin-mediated endocytosis of epidermal growth factor receptor by RNA interference. J. Biol. Chem. 279, 16657–16661. Poupon, V., Girard, M., Legendre-Guillemin, V., Thomas, S., Bourbonniere, L., Philie, J., Bright, N.A., and McPherson, P.S. (2008). Clathrin light chains function in mannose phosphate receptor trafficking via regulation of actin assembly. Proc. Natl. Acad. Sci. USA 105, 168–173. Eisenberg, E., and Greene, L.E. (2007). Multiple roles of auxilin and hsc70 in clathrin-mediated endocytosis. Traffic 8, 640–646. Holstein, S.E., Ungewickell, H., and Ungewickell, E. (1996). Mechanism of clathrin basket dissociation: separate functions of protein domains of the DnaJ homologue auxilin. J. Cell Biol. 135, 925–937. Lee, D.W., Wu, X., Eisenberg, E., and Greene, L.E. (2006). Recruitment dynamics of GAK and auxilin to clathrin-coated pits during endocytosis. J. Cell Sci. 119, 3502–3512. Massol, R.H., Boll, W., Griffin, A.M., and Kirchhausen, T. (2006). A burst of auxilin recruitment determines the onset of clathrin-coated vesicle uncoating. Proc. Natl. Acad. Sci. USA 103, 10265–10270. Schmid, S.L., Braell, W.A., Schlossman, D.M., and Rothman, J.E. (1984). A role for clathrin light chains in the recognition of clathrin cages by ‘uncoating ATPase’. Nature 311, 228–231. Ungewickell, E., Ungewickell, H., Holstein, S.E., Lindner, R., Prasad, K., Barouch, W., Martin, B., Greene, L.E., and Eisenberg, E. (1995). Role of auxilin in uncoating clathrin-coated vesicles. Nature 378, 632–635. Kelly, E., Bailey, C.P., and Henderson, G. (2008). Agonist-selective mechanisms of GPCR desensitization. Br. J. Pharmacol. 153, S379– S388. Moore, C.A., Milano, S.K., and Benovic, J.L. (2007). Regulation of receptor trafficking by GRKs and arrestins. Annu. Rev. Physiol. 69, 451–482. Premont, R.T., and Gainetdinov, R.R. (2007). Physiological roles of G protein-coupled receptor kinases and arrestins. Annu. Rev. Physiol. 69, 511–534. McPherson, J., Rivero, G., Baptist, M., Llorente, J., Al-Sabah, S., Krasel, C., Dewey, W.L., Bailey, C.P., Rosethorne, E.M., Charlton, S.J., et al. (2010). m-opioid receptors: correlation of agonist efficacy for signalling with ability to activate internalization. Mol. Pharmacol. 78, 756–766. Puthenveedu, M.A., and von Zastrow, M. (2006). Cargo regulates clathrin-coated pit dynamics. Cell 127, 113–124. Mundell, S.J., Luo, J., Benovic, J.L., Conley, P.B., and Poole, A.W. (2006). Distinct clathrin-coated pits sort different G protein-coupled receptor cargo. Traffic 7, 1420–1431. Hill, B.L., Drickamer, K., Brodsky, F.M., and Parham, P. (1988). Identification of the phosphorylation sites of clathrin light chain LCb. J. Biol. Chem. 263, 5499–5501. Wilde, A., and Brodsky, F.M. (1996). In vivo phosphorylation of adaptors regulates their interaction with clathrin. J. Cell Biol. 135, 635–645. Semerdjieva, S., Shortt, B., Maxwell, E., Singh, S., Fonarev, P., Hansen, J., Schiavo, G., Grant, B.D., and Smythe, E. (2008). Coordinated regulation of AP2 uncoating from clathrin-coated vesicles by rab5 and hRME-6. J. Cell Biol. 183, 499–511. Korolchuk, V.I., and Banting, G. (2002). CK2 and GAK/auxilin2 are major protein kinases in clathrin-coated vesicles. Traffic 3, 428–439. Olusanya, O., Andrews, P.D., Swedlow, J.R., and Smythe, E. (2001). Phosphorylation of threonine 156 of the m2 subunit of the AP2 complex is essential for endocytosis in vitro and in vivo. Current Biology 11, 896–900.

Current Biology Vol 22 No 15 1370

31. Lee, S.Y., Wenk, M.R., Kim, Y., Nairn, A.C., and De Camilli, P. (2004). Regulation of synaptojanin 1 by cyclin-dependent kinase 5 at synapses. Proc. Natl. Acad. Sci. USA 101, 546–551. 32. Bodmer, D., Ascan˜o, M., and Kuruvilla, R. (2011). Isoform-specific dephosphorylation of dynamin1 by calcineurin couples neurotrophin receptor endocytosis to axonal growth. Neuron 70, 1085–1099. 33. Watts, C. (1985). Rapid endocytosis of the transferrin receptor in the absence of bound transferrin. J. Cell Biol. 100, 633–637. 34. Traub, L.M. (2009). Tickets to ride: selecting cargo for clathrin-regulated internalization. Nat. Rev. Mol. Cell Biol. 10, 583–596. 35. Girard, M., Allaire, P.D., McPherson, P.S., and Blondeau, F. (2005). Non-stoichiometric relationship between clathrin heavy and light chains revealed by quantitative comparative proteomics of clathrin-coated vesicles from brain and liver. Mol. Cell. Proteomics 4, 1145–1154. 36. Hanyaloglu, A.C., and von Zastrow, M. (2008). Regulation of GPCRs by endocytic membrane trafficking and its potential implications. Annu. Rev. Pharmacol. Toxicol. 48, 537–568. 37. Pippig, S., Andexinger, S., and Lohse, M.J. (1995). Sequestration and recycling of beta 2-adrenergic receptors permit receptor resensitization. Mol. Pharmacol. 47, 666–676.