Traffic 2005; 6: 474–487 Blackwell Munksgaard
Copyright
#
Blackwell Munksgaard 2005
doi: 10.1111/j.1600-0854.2005.00292.x
An Extended Tyrosine-Targeting Motif for Endocytosis and Recycling of the Dense-Core Vesicle Membrane Protein Phogrin Christina Wasmeier*, Patricia V. Burgos, Tammy Trudeau, Howard W. Davidson and John C. Hutton† Barbara Davis Center for Childhood Diabetes, University of Colorado at Denver and Health Sciences Center, 4200 East 9th Avenue, Box B140, Denver, CO 80262, USA † Corresponding author: John C. Hutton,
[email protected] Integral membrane proteins of neuroendocine densecore vesicles (DCV) appear to undergo multiple rounds of exocytosis; however, their trafficking and site of incorporation into nascent DCVs is unclear. Previous studies with phogrin (IA-2b) identified sorting signals in the luminal domain that is cleaved posttranslationally; we now describe an independent DCV targeting motif in the cytosolic domain that may function at the level of endocytosis and recycling. Pulsechase radiolabeling and cell surface biotinylation experiments in the pituitary corticotroph cell line AtT20 showed that the mature 60/65 kDa form that resides in the DCV is generated by limited proteolysis in a post-trans Golgi network compartment with similar kinetics to the formation of the principal cargo, ACTH. Phogrin is exposed on the cell surface in response to stimuli and progressively internalized to a perinuclear compartment that overlaps with recycling endosomes marked by transferrin. Chimeric molecules of phogrin transmembrane and cytosolic sequences with the interleukin-2 receptor a chain (Tac) were sorted to DCVs through the action of an extended tyrosine-based motif Y654QELCRQRMA located in a 27aa sequence adjacent to the membranespanning domain. A 36aa domain terminating in this sequence conferred DCV localization to Tac in the absence of any other cytosolic or luminal phogrin components. The endocytosis and DCV targeting of phogrin Y654 > A mutants correlated with the impaired binding of the phogrin cytosolic tail to the m-subunit of the AP2 adaptor complex in vitro. Key words: dense-core granule, membrane traffic, neuroendocrine, regulated secretion, sorting signal Received 5 December 2002, revised and accepted for publication 23 March 2005, published on-line 29 April 2005
* Present address: Cell and Molecular Biology Section, Division of Biomedical Sciences, Imperial College, Sir Alexander Fleming Building, Exhibition Road, London SW7 2AZ, UK.
474
Most polypeptide hormones and neuropeptides are stored in dense-core vesicles (DCVs or secretory granules) and released by regulated exocytosis. Biogenesis of these organelles occurs at the level of the trans Golgi network (TGN) and is characterized by the focal concentration of secreted product in the TGN, the budding of nascent granules and a series of membrane remodeling events that include homotypic fusion and clathrin-mediated excision of selected membrane and vesicular components. Resident membrane proteins of the mature granule play key roles in many of these sorting events and are also central to the subsequent interactions of the granule with the cytoskeleton and membrane fusion during granule exocytosis (1,2). Some DCV membrane proteins, such as the proton translocating vacuolar ATPase complex, are found in other intracellular vesicles such as endosomes, lysosomes and synaptic vesicles (SVs) while others show specific association with the DCV. The functional integrity of the DCV and its biochemical and biophysical activities such as prohormone conversion and the storage of biogenic amines depend on the maintenance of the membrane protein composition and function. This occurs in the face of wide variation in rates of DCV exocytosis and biogenesis, and secretagogue induced changes in the biosynthetic rates of cargo and membrane proteins. There is a paucity of information regarding the relative contributions to granule biogenesis of de novo biosynthesis of membrane components versus recycling, and questions remain regarding whether DCV membrane proteins traffic individually or collectively (3) and where in the cell they intersect with newly synthesized DCV cargo. Trafficking information concerning individual DCV integral membrane proteins is limited. The best characterized to date is peptidylglycine a-amidating monooxygenase (PAM/ PAL), a type 1 transmembrane protein that occurs in both integral membrane and soluble forms in most neuroendocrine cells. The luminal catalytic domains can be sorted autonomously, but additional signals within the cytoplasmic tail are required for the reinternalization of PAM molecules from the cell surface and the targeting of the protein to the TGN (4–6). P-selectin, a type 1 transmembrane protein which is normally found in storage granules in platelets and endothelial cells traffics to DCVs when expressed heterologously in neuroendocrine cell lines. Its 35aa cytoplasmic tail is necessary and sufficient for such targeting (7); however, this region contains overlapping sorting information that can direct the molecule to
Recycling of Phogrin
synaptic-like microvesicles (termed SVs in this article) and lysosomes (8–11). The multispanning vesicular monoamine transporters (VMAT) that can be localized to either DCVs or SVs appear to be targeted by a COOH-terminal dileucine motif (12–14). Phogrin (IA-2b), the molecule used in the present studies, is a 60–65-kDa type I integral membrane glycoprotein localized to DCVs of a wide range of neuronal and endocrine cell types (15–17). Its 376aa cytoplasmic region incorporates a 260aa protein tyrosine phosphatase (PTP) domain flanked N-terminally by a serine-rich 105aa juxtamembrane region and an 11aa COOHterminal tail. The PTP domain adopts a typical PTP fold but is catalytically inactive on the basis of enzymatic measurements and structural considerations (Schoelson and Hutton, unpublished findings). Phogrin bears the signature of a larger subfamily of secretory granule PTP members that includes mammalian IA-2 (ICA512) (18,19), Drosophila FLYDA and Caenorhabditis elegans IDA-1 (2,17). IA-2 and phogrin overlap in their cellular distribution but appear to be differentially regulated (20). The phogrin cytosolic domain is phosphorylated in a secretagoguedependent manner in pancreatic b-cells (21,22). IA-2, by contrast, undergoes m-calpain-mediated proteolytic cleavage in response to secretory stimuli (23) to produce a C-terminal fragment that is translocated to the nucleus (24). Phogrin and IA-2 show homotypic and heterotypic interactions and association with cytoskeletal elements (25) and active PTP molecules (26). Disruption of either the IA-2 or the phogrin gene in the mouse results in impaired glucose tolerance (27,28) consistent with a role in the regulation of b-cell insulin secretion. IDA-1 mutants in C. elegans exhibit a defective egg-laying phenotype similar to that observed in mutants deficient in certain neuropeptides and neuropeptide-processing enzymes (29). The current report investigates the identity of targeting signals in the conserved cytosolic domain of phogrin using stably transfected AtT20 corticotroph cells as a model system. A series of mutant forms of phogrin (Supplement 1 available on-line at http://www.traffic.dk/suppmat/6_6b.asp) were generated based on either the full length sequence or the chimeric constructs in which the luminal and transmembrane sequence was substituted by the corresponding sequences in the interleukin-2 (IL-2) receptor (Tac), a molecule that is excluded from the DCV compartment (5,30,31). We demonstrate that a DCV targeting sequence in the phogrin cytoplasmic region based on a tyrosine-based motif with a 6aa extension is sufficient to direct Tac chimeras to DCVs. We conclude that clathrin-mediated endocytosis and recycling of phogrin from the plasma membrane are important in maintaining its steady-state localization.
Results Post-translational processing and intracellular storage of phogrin in AtT20 cells AtT20 anterior pituitary cells express low levels of phogrin endogenously (32) making them a suitable system for Traffic 2005; 6: 474–487
studying the targeting of mutant forms of the molecule in a neuroendocrine cell type without the need of epitope tags. The level of expression of the transfected phogrin constructs exceeded that of the endogenous molecule by more than fivefold as ascertained by Western blotting (32) and radiolabeling (data not shown) of stably transfected versus mock-transfected cells. This makes it unlikely that trafficking of any mutant versions of the molecule would be influenced by interactions with endogenous wild-type molecules. Radiolabeling experiments were routinely performed using a 2-h labeling protocol at 20 C aimed at maximizing radioisotope incorporation while arresting exit of the molecule from the TGN. Under these conditions, the major immunoprecipitable form of phogrin was 130 kDa, a size identical to phogrin generated by labeling at 37 C for 20 min or in in vitro translation in the presence of dog pancreatic microsomes, and corresponding to the glycosylated proprotein form (15). A 3-h chase at 37 C after labeling at 20 C resulted in the intracellular accumulation of a doublet of 60/65 kDa, a lesser amount of a 90 kDa form, and only small amounts of the 130-kDa form (Supplement 2 available on-line at http://www.traffic.dk/ suppmat/6_6b.asp). The conversion of the 130-kDa form was blocked by 1 mg/mL of brefeldin A (Supplement 2 available on-line at http://www.traffic.dk/suppmat/ 6_6b.asp) which disrupts Golgi transport and by the membrane permeant base chloroquine (100 mM), which together with the block at 20 C, is consistent with the processing of phogrin occurring in a post-TGN acidic compartment, presumably the DCV. Radiolabeling and immunoprecipitation experiments were combined with cell-surface biotinylation and avidin affinity chromatography to determine the kinetics of the posttranslational modification of phogrin and gain insight into the trafficking of phogrin to the plasma membrane (Figure 1). After a delay of approximately 30 min, the 130-kDa form of phogrin progressively disappeared which was accompanied by the reciprocal appearance of the 60-kDa and 90-kDa forms indicative of a precursor – product relationship (Figure 1A,B). Conversion of the precursor followed first order kinetics with a t1/2 of 50–60 min and was virtually complete by 3 h (Figure 1D,E). The molecular composition of immunoprecipitable phogrin changed little between 2 and 6 h of chase, with the 60 and 90-kDa products declining in parallel (t1/2 ¼ 8–11 h) (data not shown). At the cell surface, newly synthesized phogrin appeared principally as the processed 60 and 90-kDa forms between 30 and 60 min post release of the 20 C block (Figure 1A,C). The 130-kDa precursor was poorly represented at the surface relative to the intracellular pool at all time-points consistent with exocytosis occurring after initial DCV maturation (33). The transferrin receptor in the same cells appeared more rapidly at the cell surface ( A substitution in the context of an otherwise unmodified cytosolic tail, T-T-P/ Y654A, was created. In contrast to T-T-P/nPTP, this 480
protein was not targeted to the tips of cell extensions and did not show co-localization with ACTH in other parts of the cell (Figure 6) but instead was associated with the cell surface and a perinuclear compartment similar to T-T-P/nJM. Surprisingly, T-T-P that was truncated immediately following the native tyrosine-based signal (T-T-P/YQEL.tr) was also excluded from the tips but exhibited less surface labeling than T-T-P/Y654A and under steadystate conditions was largely found in the TGN as indicated by its co-localization with TGN38 (data not shown). An alignment of members of the phogrin/IA-2 family across a broad range of species (Supplement 5 available on-line at http://www.traffic.dk/suppmat/6_6.asp) indicated that the YXXj motif in most instances is flanked C-terminally by a consensus CRQ sequence (aa 659–661). When these residues were substituted with Ala in the context of an Traffic 2005; 6: 474–487
Recycling of Phogrin T-T-P/∆PTP
T-T-P/∆JM
T-T-P/Y654A
T-T-P/YQEL.tr YQEL*
Merge
ACTH
Tac
AQEL
Figure 6: Subcellular localization of mutant forms of T-T-P. AtT20 cells were transiently (T-T-P/nPTP and T-T-P/nJM) or stably (T-TP/Y654A and T-T-P/YQEL.tr) transfected with constructs based on T-T-P containing mutations within the cytosolic region. Cells were fixed, permeabilized and stained as described in Figure 5. Bar ¼ 10mm.
otherwise wild-type cytosolic construct (T-T-P/ CRQ > AAA, Supplement 1 available on-line at http:// www.traffic.dk/suppmat/6_6.asp), the protein was also excluded from ACTH-positive cell processes (Figure 7). A molecule truncated immediately after the CRQ sequence (T-T-P/CRQ) was also essentially excluded from cell extensions; however, inclusion of a further three residues (T-T-P/RMA*) restored efficient trafficking to DCVs. We conclude that the sequence CRQ in the context of an extended tyrosine-based motif represents a sorting element contributing to secretory granule targeting of phogrin. PhogrinY654A accumulates at the cell surface The data presented above strongly implicate the YQEL motif in the trafficking of phogrin lacking its luminal domain. To confirm that this is also the case in the context of the native molecule, a construct was made in which tyrosine654 was mutated (Y654 > A). As predicted, this molecule showed increased membrane staining in unpermeabilized cells and reduced intracellular localization in the tips of the cells relative to the control (Figure 8A, compare with Figure 3), although less-concentrated perinuclear staining than the ncyto construct. Interaction of phogrin with adaptor complexes Tyrosine-based motifs typically recruit cytosolic coat complexes of the hetero-tetrameric adaptor family through binding to their medium chain (m) subunits (39–42). The 130 amino acid juxtamembrane segment of phogrin (nPTP) fused to glutathione-S-transferase (GST) (Supplement 1) was tested for interaction with the m chains of the AP1, AP2, AP3 and AP4 complexes. GST/ nPTP showed specific binding to in vitro translated [35S]Traffic 2005; 6: 474–487
labeled m2 (Figure 8B). Binding of in vitro translated m1, m3 and m4 to GST/nPTP was not significantly higher than GST alone ( A construct observed by immunofluorescence microscopy above (Figure 8A).
Discussion Accumulating evidence suggests that proteins within the membrane of polypeptide-containing DCVs are retrieved from the cell surface following stimulated exocytosis and recycled for subsequent exocytotic events (3,42–47). In this model, transmembrane DCV proteins are transiently incorporated into the plasma membrane during regulated exocytosis, and after internalization, traffic through endosomal structures that intersect with membrane limited compartments containing nascent secretory cargo at the trans Golgi or a Golgi-derived vesicle such as the immature secretory granule. It is likely that multiple modes of endocytosis operate in DCV membrane recycling depending on whether DCV components intermix with the plasmalemma or are captured as an organelle that is empty or partially depleted of contents (3,45). Aside from the mechanics, there is the question of the extent to which individual DCV membrane proteins can traffic 481
Wasmeier et al. T-T-P/RMA.tr
T-T-P/CRQ.tr YQELCRQ*
YQELAAA
Merge
α−ACTH
α−Tac
YQELCRQRMA*
T-T-P/CRQ→AAA
Figure 7: Mapping of DCV targeting sequences in T-T-P to an extended tyrosine motif. AtT20 cells were transiently transfected with constructs encoding the indicated T-T-P deletion, point or truncation mutants, and analyzed by immunofluorescence microscopy as described in Figure 5. Bar ¼ 10mm.
independently of, or in association with, other DCV components to reach their destination (44). As many components of the DCV membrane are shared with other endosomal and lysosomal structures, the possibility arises that different itineraries are taken by different components. Phogrin, the DCV membrane marker protein, used in these studies has proven to be a robust marker of these organelles and has been used to track DCV movements along microtubular pathways (2,16) and to monitor membrane fusion during exocytosis of DCV cargo, the intragranular pH, and the release of DCV (3,45). Total internal reflectance fluorescence microscopy (TIRF) reveals that phogrin-GFP integrates into the plasma membrane following cell stimulation but disperses slowly from the site of exocytosis relative to other DCV markers such as VAMPGFP (45,46). It is internalized over a relatively long time– course (seconds) compared with DCV fusion, a finding that is in accord with membrane capacitance measurements which indicate that endocytic rates lag behind exocytic rates resulting in a net accumulation of DCV membrane in the plasmalemma (44). The TIRF data, however, poses 482
difficulties to interpretation due to artifacts relating to the structure of reporter constructs (46). Pulse-chase radiolabeling experiments performed here in AtT20 corticotrophs suggested that newly synthesized phogrin, which undergoes post-translational limited proteolysis in an acidic environment, reaches the cell surface by essentially the same pathway as secreted-DCV cargo, namely via the TGN and immature secretory granule intermediate. Phogrin traffic could be distinguished kinetically from that of the transferrin receptor, which is likely to use a constitutive pathway of Golgi to plasma membrane transport, possibly via endosomes (48). Exposure of the cells to a secretory stimulus increased surface exposure of the mature 60 and 90-kDa forms of phogrin, which appeared to be countered by a relatively slow rate of endocytosis compared with that of the transferrin receptor. The increase of plasmalemmal phogrin in response to secretory stimuli we observed by immunofluorescence microscopy and surface biotinylation experiments is consistent with the steady-state capacitance recordings in stimulated b-cells (44). Under basal conditions, the cellular content of phogrin turned over slowly (t1/2 8–10 h), yet a Traffic 2005; 6: 474–487
Recycling of Phogrin A
Surface
Cellular Y654A mutant
B 30
Binding (%)
25 20 15 10 5 0 GST
YQEL
AQEL
AQEA
Figure 8: The phogrin cytosolic domain interacts with AP2m chains. A) AtT20 cells stably transfected with phogrin Y654A were either incubated at 4 C with rabbit anti-phogrin N-terminal without permeabilization and then fixed (surface binding, left panels) or first fixed and permeabilized before reaction with antibody (cellular, right panels). Bound antibody was detected with Cy2-conjugated anti-rabbit IgG, and cells were analyzed by immunofluorescence microscopy. See Figure 3 for control incubation. B) The indicated recombinant forms of phogrin cytoplasmic domain fused to GST were incubated with in vitro translated radioactively labeled m2 subunit. Complexes were isolated using glutathione – agarose beads, separated by SDS PAGE in parallel with an aliquot of the starting material and analyzed by phosphorimaging. Bound radioactivity is expressed as a proportion of the starting material (SEM, n ¼ five experiments).
significant proportion of the molecules were transiently exposed on the cell surface (15–20%). Given that the surface pool turns over every 20–30 min, this implies that a large proportion of the plasmalemmal pool is recycled. Internalization of phogrin from the cell surface appeared to be linked to clathrin-mediated endocytosis as evidenced by the dependence on a classical YXXj motif within the cytoplasmic tail and an association with the AP2 adaptor complex. Similar YXXj motifs are found in other proteins Traffic 2005; 6: 474–487
that are associated with the DCV membrane, namely peptide amidating monooxygenase (5), P-selectin (10) and VMAT1 (49); however, this may reflect the generic nature of this signal as a mechanism to recruit coat proteins for traffic (40) as opposed to a specific endocytic or DCV targeting mechanism. A case in point here is the transferrin receptor that was internalized more rapidly than phogrin and yet shares a common destination as indicated by the delivery of internalized anti-phogrin antibodies to recycling endosomes. The differences in internalization kinetics between phogrin and the transferrin receptor suggest mechanistic differences, but conceivably relates to phogrin entry as part of a macromolecular complex, or its interactions with cytoskeletal elements such as b2 syntrophin and b4 spectrin (23,25) that could impede its progress. The delivery of phogrin to recycling endosomes in AtT20 cells observed in these experiments, however, is at odds with similar studies performed in the pancreatic endocrine cell lines Min6 and INS-1 that demonstrated delivery of phogrin-bound antibodies from the cell surface to DCVs via a syntaxin-6-positive compartment (47,50) that may represent the immature secretory granule (51). It is possible that this discrepancy reflects the fact that in AtT20 cells, unlike the b-cell lines, the phogrin antibodies dissociated from the antigen in an endosomal compartment and hence do not accurately reflect the antigen’s final destination. In support of this notion, antibodies (guinea pig) directed at the COOH terminus of phogrin applied to permeabilized cells that had internalized luminally directed rabbit anti-phogrin antibodies did not co-localize in the transferrin receptor-positive compartment and yet decorated DCVs strongly in the tips of the cells (data not shown). Antibody internalization experiments performed in AtT20 cells that investigated the endocytosis of PAM similarly showed delivery to recycling endosomes (4) although this is not postulated as the final destination of the transmembrane form of the enzyme (6). This might suggest that the recycling endosome is a potentially important point of intersection between the endocytic and regulated secretory pathways, as well as with the constitutive secretory pathway that delivers proteins to the cell surface (52). The single YXXj motif in the cytosolic tail of phogrin that is implicated in its internalization via AP2 is likely to function in the further itinerary of the molecule. Thus, the membrane-associated form of PAM has a related YXXj motif (Y968SRP) close to its C-terminus that operates in concert with the surrounding 18aa residues to promote its retention in a non-endosomal, tubulovesicular compartment that overlaps with the TGN (5,6). The signal is dominant to the luminal PAM domain that can direct the molecule to the DCV from the TGN. Similarly, P-selectin has a Y777GVF sequence in its short cytoplasmic tail which, when mutated, affects its internalization from the plasma membrane and acts in concert with the sequence L768N upstream to target the molecule to DCVs in AtT20 cells (9). Sorting is depicted as occurring at the 483
Wasmeier et al.
plasmalemma but could equally involve endocytic delivery to the TGN and sorting to DCVs from that location (9). Other studies indicate that the P-selectin tyrosine motif acts in concert with K765CPL 8aa upstream to direct it from early endosomes to SVs (11) or with a D786PSP motif 5aa downstream to direct traffic from late endosomes to SVs (11). Our data with phogrin indicate that the tyrosine motif YQEL, with a COOH-terminal extension of six amino acids, is required to target a chimeric construct of the cytosolic region of phogrin and luminal and transmembrane domains of the plasma membrane protein Tac to DCVs in AtT20 cells. The motif (YQELCRQRMA in rat phogrin) is strikingly conserved from man to C. elegans in the larger family of PTP molecules that localize to DCVs with the consensus sequence YQ (D,E) LCR (Q,A) (R,H) MA (Supplement 5 available on-line at http://www.traffic. dk/suppmat/6_6b.asp). Substitution of CRQ with AAA in the context of the full-length cytosolic domain prevented localization of the mutant protein to DCVs, suggesting that at least one of these residues forms part of a granule targeting motif in concert with the YXXj sequence. In light of this and previous studies, further mutagenesis studies are warranted to further delineate the critical residues within the region and the NH2-flanking sequence. The Tac-phogrin chimera truncated after the extended tyrosine-based signal (YQELCRQRMA), while efficiently targeted to DCVs, does not precisely recapitulate the intracellular localization of native phogrin. Compared with cells transfected with wild-type phogrin, the cells expressing the truncated chimera show increased perinuclear and plasma membrane labeling and a greater number of cytoplasmic vesicles that are ACTH negative. There was, however, no evidence for targeting of such Tac-phogrin chimeras to SVs, lysosomes or endosomes, which would have suggested diversion in the itinerary of phogrin. It is most likely therefore that the YQELCRQRMA motif acts synergistically with additional sorting motifs within the luminal domain of phogrin, which itself can direct sorting of phogrin to the DCV independently of the remainder of the molecule via as-yet poorly defined motifs principally in the proregion (aa 31–415) (32). It should be noted, however, that the luminal domain is cleaved after sorting to DCVs. At present, the fate of the cleaved prosequence is uncertain, but it is reasonable to suggest that it is secreted during exocytosis and thus not available to direct subsequent trafficking of the molecule post endocytosis. It is a formal possibility that the prosequence is retained through a non-covalent interaction with the mature NH2 terminus of the molecule; however, there was little evidence of this based on co-immunoprecipitation experiments (data not shown). As some targeting information resides within the ‘mature’ portion of the luminal domain (32), this may be responsible for the increased efficiency of targeting of wild-type constructs relative to the truncated chimeras, although the presence of additional subdominant determinants within the transmembrane and/or cytoplasmic domains of the native molecule cannot be excluded. 484
In addition to the extended tyrosine-based motif, the phogrin COOH terminus also contains a conserved dileucinelike motif (E992EVNAILKALPQ1003) that is reminiscent of both (D,E) XXXL (L,I) sequences that mediate binding to AP subunits, and acidic cluster dileucine sequences (EEXXLL) involved in the recruitment of GGAs (53). The sequence is highly conserved from mammals to teleost and cartilaginous fishes in the phogrin/IA-2 family and has been implicated in the targeting of VMATs to DCVs (12–14,49) (Supplement 5 available on-line at http://www. traffic.dk/suppmat/6_6b.asp). In the current study, a chimeric molecule of Tac with the phogrin cytosolic domain truncated at residue 758 was efficiently targeted to DCVs, arguing against the dileucine-like motif acting as an autonomous DCV-sorting signal. However, we cannot currently preclude that it enhances the efficiency of sorting, for example by acting in concert with the tyrosine-based motif through adaptor molecules such as AP3 (53) or by promoting interactions with the cytoskeleton (27). Our current working hypothesis is that newly synthesized phogrin is routed into the regulated pathway of secretion at the level of the TGN largely on the basis of sorting information within the propeptide (32) which apparently acts in a dominant fashion. DCV exocytosis delivers phogrin to the cell surface, and like a number of other granule membrane components (5,19,42,43), it is subsequently retrieved and recycled. The single tyrosine-based motif located with the 365aa cytosolic tail appears to promote re-entry of phogrin into the cell through clathrin-dependent AP2-mediated endocytosis. We postulate that a 36aa segment of the cytosolic domain adjacent to the DCV membrane that contains this motif is sufficient to target phogrin back to DCVs but most probably operates in combination with other linear motifs and/or post-translational modifications to achieve efficient targeting. Precisely where in the cell these sorting motifs operate is presently unclear and is the subject of on-going investigations. It should also be borne in mind that phogrin interacts with a number of DCV and cytoskeletal proteins and may traffic as a multimolecular complex (25,26). While it may provide a vehicle for the recycling of other DCV components, it may equally receive additional sorting instructions from other DCVassociated proteins.
Materials and Methods Reagents All chemicals were purchased from Sigma (St Louis, MO, USA) or Fisher Scientific (Pittsburg, PA, USA) unless indicated. Molecular biology reagents were from New England Biolabs (Beverly, MA, USA) and cell-culture supplies were obtained from Invitrogen (Carlsbad, CA, USA).
CDNA constructs pCDNA3/phogrin and soluble luminal domain were described previously (15,21,32). Mutants (see supplement 1) were generated by polymerase chain reaction with Pfu polymerase (Stratagene, La Jolla, CA, USA) (21), using the primers listed in combination with appropriate vector-specific
Traffic 2005; 6: 474–487
Recycling of Phogrin primers. All constructs were finally subcloned into pCDNA3 (Invitrogen). Phogrin n cytoplasmic tail has a predicted 13aa cytoplasmic extension (RHRSSMHLEGPILstop), comprising the phogrin RHR stop-transfer sequence and 10 randomly selected residues to ensure membrane anchoring and was created with the reverse primer 5´GTAGCTCGAGCGGTGGCGGAGGCAGTAGGCTAG (mutations underlined). Tac was excised from pCDM8/Tac [a gift from Dr J. Bonifacino, National Institutes of Health (NIH)] with EcoRI and XbaI and subcloned into pcDNA3 to generate pcDNA3/Tac. To create the T-P-P chimeric construct, a BglII site was introduced at the 3´ end of the Tac luminal domain (5´AGATCTGTTGTAAATATGGACGTCTCC). The phogrin cytosolic/ transmembrane region was amplified using 5´AGATCTCGACAA GTTCATTGTGCTTACCTTC. For T-T-P and deletion mutants, a SacII site was introduced at the 3´ end of the Tac transmembrane span (5´CCGCGGGAGCCCACTCAGGAGGAGG). The phogrin-cytosolic domain was amplified with 5´CCGCGGCACAACTCACACTACAAGC. C-terminally truncated cytosolic domains were created using 5´GTTCTCTAGAG CCTACTCCTCTCTCTGGGCCACCAG for T-T-P/nPTP, 5´CTAACCTCCG CCGCCTAGCTCCTGGTAGGCTTCGG for T-T-P/YQEL, 5´CTAACCTCC GCCGCCTTGGCGGCATAGCTCCTGGTAG for T-T-P/CRQ and 5´CTAAC CTCCGCCGCCAGCCATACGTTGGCGGCATAG for T-T-P/RMA (note that for molecules truncated immediately after the extended tyrosinebased motif, a C-terminal extension of four glycine residues was added) (54). For internal deletions, the following primers were used: 5´GTAGCTCGAGCGGTGGCGGAGGCAGTAGGCTAG and 5´CAGAGCTCGA GGAATGCACCCAAGAACCGTTCCC3´ for T-T-P/nJM; 5´AGATCTGAT CCTGCGGCCGAACAGC and 5´AGATCTCACTGTGCGCCTATCAAGCA for Point mutations were introduced using T-T-P/n669-741. 5´TGCCACCGAAGCCGCCCAGGAGCTATG/5´GGCATAGCTCCTGGGCGGC TTCGGTGGC for T-T-P/Y654A and 5´TACCAGGAGCTAGCCGCCGCA CGTATGGCTGTTCGG/5´AACAGCCATACGTGCGGCGGCTAGCTCCTGGTA GGC for T-T-P/CRQ!AAA. pEYFP-rhoB was purchased from BD Biosciences (San Diego, CA, USA) and Clontech (Palo Alto, CA, USA).
Cell culture and transfection AtT20 mouse anterior pituitary cells were grown in Dulbecco’s modified Eagle’s medium/F12 HAM base (3:1 ratio) supplemented with 10% (v/v) donor calf serum with iron, 10% horse serum, 2% fetal bovine serum and antibiotics. Cells were transfected with LipofectAMINE 2000 reagent (Invitrogen) according to the manufacturer’s recommendations. Stably expressing cell lines were selected initially with 1 mg/mL of G418 and maintained in medium containing 0.25 mg/mL of G418. Western blotting and immunofluorescence microscopy was used to select moderately expressing clones for study. The reported results were replicated in 3–4 independently selected clones. Transiently expressing cells were transfected in the same way and fixed after 48 h. In this case and in the instance of pEGFP-RhoB constructs, cells with low-to-moderate levels of expression were chosen to avoid potential artifacts arising from perturbation of normal membrane trafficking. Experimental manipulations were performed following preincubation of cells in serum-free culture media containing 1% BSA (basal media) for 1 h followed by incubation in the same media with specific modifications as noted below. Exocytosis was stimulated either by exposure of cells to 1 mM BaCl2 in Ca-free basal media or in basal media containing 55 mM KCl and 10 mM forskolin.
Pulse-chase radiolabeling and surface biotinylation Transfected AtT20 cells were grown as above to 70% confluency then preincubated in serum-free Dulbecco’s modified Eagle’s medium depleted of methionine and cystine for 1 h at 37 C. 35S-labeled amino acid mixture (Amersham Biosciences, Piscataway, NJ, USA) was added to each culture (100 mCi in 1 mL volume in a 3.5-cm Petri dish), and the cells were incubated at 20 C for 2 h. After washing to remove unincorporated label, complete media was added and chase incubations of up to 6 h conducted at 37 C. Subsequent steps were conducted at 4 C. Cells were then rinsed twice in PBS and exposed to 1 mM Sulfo-NHS-LC Biotin reagent (Pierce Biotechnology, Rockford, IL, USA) in PBS for 30 min. The cells were subsequently washed, incubated in 50 mM NH4Cl for 10 min to
Traffic 2005; 6: 474–487
quench un-reacted ester, rinsed twice more with PBS then lysed in 150 mM NaCl, 50 mM Tris pH 7.5, 1% (v/v) Triton X-100 and immunoprecipitated with anti-phogrin antibodies. Immunocomplexes were recovered with protein A agarose, bound proteins eluted in 30 mL of 0.5% SDS and 1% b-mercaptoethanol and then diluted either with an equal volume of 2X SDS-PAGE loading buffer (total incorporation) or with 1 mL of lysis buffer for binding to neutravidin beads (Pierce Biotechnology). After SDS PAGE, gels were dried and quantitated by Phosphor-imaging (Storm; Molecular Dynamics, Sunnyvale, CA, USA). Unlabeled cells stimulated for 15 min with 1 mM BaCl2 were rinsed twice in PBS at 4 C and treated for 30 min with 1 mM Sulfo-NHS-SS-Biotin reagent (Pierce Biotechnology) in PBS. The cells were subsequently washed, incubated in 50 mM NH4Cl for 10min, rinsed twice in PBS and subsequently incubated for the indicated times in complete media at 37 C. Surface biotin was removed by incubation of the cells at 4 C with 50 mM glutathione for 30 min followed by 50 mM iodoacetamide to quench free sulfhydryls. The cells were then lysed in 50 mM Hepes buffer pH 7.4 containing 150 mM NaCl, 10% (v/v) glycerol, 1% Triton X-100 and a protease inhibitor cocktail comprised of 2.5 mM iodoacetamide, 1 mM PMSF, 1 mg/mL of pepstatin and 1 mg/mL of E-64. Biotinylated proteins were captured on neutravidin beads (Pierce Biotechnology) and then fractionated by SDS PAGE and detected by Western blotting performed with an Nterminally directed anti-phogrin antibody (Cyrus) or commercially available control antibodies using enhanced chemiluminescence as a detection system (Amersham Biosciences).
Immunofluorescence microscopy Cells grown on glass coverslips were rinsed in PBS, fixed for 30 min in 4% paraformaldehyde in PBS and permeabilized with 0.1% Triton X-100 in PBS. All blocking, labeling and washing, steps were carried out in PBS/ 0.2% BSA. The following antibodies were used: affinity-purified rabbit antiphogrin luminal domain (15); rabbit anti-Tac (R3134, from Dr J. Bonifacino); mouse monoclonal anti-Tac (clone 143–13, Biosource, Camarillo, CA, USA); mouse anti-ACTH (clone A-2A3, from Dr A. White, Manchester, UK); mouse antisynaptophysin (clone SY38, Chemicon International, Temecula, CA, USA); mouse anti-syntaxin 6 (clone 30, BD Biosciences); rat anti-LAMP-1 (1D4B, Developmental Studies Hybridoma Bank, Iowa City, IA, USA); rabbit or sheep anti-TGN38 (from Paul Luzio, Cambridge UK and Serotec, Raleigh, NC, USA, respectively). Labeling was visualized using Cy2-conjugated anti-rabbit, rhodamine-conjugated anti-rabbit, Cy3conjugated anti-mouse or antisheep or FITC-conjugated anti-rat secondary antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA, USA). Coverslips were mounted using Mowiol mounting medium (Calbiochem, San Diego, CA, USA). Images were acquired with a 60 oil lens using a Nikon Microphot-FXA microscope equipped with a Micromax 1-Hz digital camera (Roper Scientific, Tucson, AZ, USA) and SLIDEBOOK software (Intelligent Imaging Innovations, Denver, CO, USA). For digital deconvolution, a series of Z-sections at 0.2-mm intervals was acquired and deconvolved using a ‘nearest neighbors’ algorithm in SLIDEBOOK. For analysis of transient transfection experiments, cells exhibiting moderate expression levels similar to those seen in stably transfected lines were chosen for imaging.
Immunoblotting Cells were rinsed in PBS, scraped into PBS-containing protease inhibitors (10 mM E64, 10 mM pepstatin A and 1 mM phenylmethylsulfonyl fluoride) and homogenized by sonication. Proteins (30 mg/lane) were separated by SDS PAGE, transferred to nitrocellulose membranes and detected by immunoblotting using enhanced chemifluorescence as described previously (21). The following antibodies were used: affinity-purified rabbit anti-phogrin cytoplasmic domain (15) and rabbit anti-Tac (IL-2Ra) N-terminus from Santa Cruz Biotechnology (Santa Cruz, CA, USA).
GST pulldown experiments Full-length clones for the m subunits of adaptor complexes AP1, AP2, AP3 and AP4 were kindly provided by Dr J. Bonifacino (NIH, Bethesda, MD, USA) and subcloned into pCDNA3. One mg of plasmid was transcribed and translated in vitro in 50 mL in the presence of 10 mCi [35S]-methionine
485
Wasmeier et al. (0.5 mCi/mmol; Amersham Biosciences), using the TNT-Quick capped transcription/translation system (Promega) according to the manufacturer’s protocol. The translated product (50 000 dpm) was incubated for 2 h at 4 C in 200 mL of 20 mM HEPES pH 7.4, 150 mM KCl, 10 mM MgCl2, 10% (v/v) glycerol, 0.1% Triton X-100 and 50 mg % BSA in the presence of 20 mg of GST fusion protein. Fusion proteins were generated by cloning BglII/EcoRI-flanked PCR products into the BamHI/EcoRI sites of pGEX-5X-3 (Amersham Biosciences) and were expressed and purified according to the manufacturer’s instructions. Following incubation, GST fusion proteins were isolated on glutathione – agarose beads (Amersham Biosciences), washed five times in binding buffer without BSA and bound radioactivity eluted with SDS-PAGE gel-loading buffer. Proteins were separated by SDS PAGE, gels stained with Coomassie Blue, equilibrated with 10% (v/v) glycerol, dried and radioactivity determined by phosphorimaging. Binding was expressed as a percentage of radioactivity recovered relative to the starting material. Statistical analyses were performed using Student’s t-test.
Acknowledgments Juan Bonifacino (NIH) is thanked for providing the Tac plasmid, rabbit antiTac antibody R3134, and adaptor m-chain constructs, Richard Mains (University of Connecticut) for AtT20 cells and Paul Luzio (University of Cambridge) for rabbit anti-TGN38 antibodies. The work was supported by NIH grants DK55597 and DK60861, the NIH Diabetes and Endocrine Research Center at the University of Colorado at Denver and Health Sciences Center (P30 DK57516), the JDRF and Children’s Diabetes Foundation. Christina Wasmeier and Patricia V. Burgos were the recipients of JDRFI Postdoctoral Fellowships (1998–2000 and 2004–2005, respectively). Dr Nicholas Bright (University of Cambridge) provided valuable discussion, and Ms Carrie John and Ms Heather Davis provided secretarial assistance.
References 1. Solimena M, Gerdes HH. Secretory granules: and the last shall be first. Trends Cell Biol 2003;13:399–402. 2. Zahn TR, Angleson JK, MacMorris MA, Domke E, Hutton JF, Schwartz C, Hutton JC. Dense core vesicle dynamics in Caenorhabditis elegans neurons and the role of kinesin UNC-104. Traffic 2004;5:544–559. 3. Taraska JW, Perrais D, Ohara-Imaizumi M, Nagamatsu S, Almers W. Secretory granules are recaptured largely intact after stimulated exocytosis in cultured endocrine cells. Proc Natl Acad Sci USA 2003;100:2070–2075. 4. Milgram SL, Mains RE, Eipper BA. COOH-terminal signals mediate the trafficking of a peptide processing enzyme in endocrine cells. J Cell Biol 1993;121:23–36. 5. Milgram SL, Mains RE, Eipper BA. Identification of routing determinants in the cytosolic domain of a secretory granule-associated integral membrane protein. J Biol Chem 1996;271:17526–17535. 6. Milgram SL, Kho ST, Martin GV, Mains RE, Eipper BA. Localization of integral membrane peptidylglycine alpha-amidating monooxygenase in neuroendocrine cells. J Cell Sci 1997;110:695–706. 7. Disdier M, Morrissey JH, Fugate RD, Bainton DF, McEver RP. Cytoplasmic domain of P-selectin (CD62) contains the signal for sorting into the regulated secretory pathway. Mol Biol Cell 1992;3:309–321. 8. Norcott JP, Solari R, Cutler DF. Targeting of P-selectin to two regulated secretory organelles in PC12 cells. J Cell Biol 1996;134:1229–1240. 9. Modderman PW, Beuling EA, Govers LA, Calafat J, Janssen H, von dem Borne AE, Sonnenberg A. Determinants in the cytoplasmic domain of P-selectin required for sorting to secretory granules. Biochem J 1998;336:153–161.
486
10. Blagoveshchenskaya AD, Hewitt EW, Cutler DF. A complex web of signal-dependent trafficking underlies the triorganellar distribution of P-selectin in neuroendocrine PC12 cells. J Cell Biol 1999;145: 1419–1433. 11. Blagoveshchenskaya AD, Cutler DF. Sorting to synaptic-like microvesicles from early and late endosomes requires overlapping but not identical targeting signals. Mol Biol Cell 2000;11:1801–1814. 12. Tan PK, Waites C, Liu Y, Krantz DE, Edwards RH. A leucine-based motif mediates the endocytosis of vesicular monoamine and acetylcholine transporters. J Biol Chem 1998;273:17351–17360. 13. Krantz DE, Waites C, Oorschot V, Liu Y, Wilson RI, Tan PK, Klumperman J, Edwards RH. A phosphorylation site regulates sorting of the vesicular acetylcholine transporter to dense core vesicles. J Cell Biol 2000;149:379–396. 14. Waites CL, Mehta A, Tan PK, Thomas G, Edwards RH, Krantz DE. An acidic motif retains vesicular monoamine transporter 2 on large dense core vesicles. J Cell Biol 2001;152:1159–1168. 15. Wasmeier C, Hutton JC. Molecular cloning of phogrin, a protein tyrosine phosphatase homologue localized to insulin secretory granule membranes. J Biol Chem 1996;271:18161–18170. 16. Pouli AE, Emmanouilidou E, Zhao C, Wasmeier C, Hutton JC, Rutter GA. Secretory-granule dynamics visualized in vivo with a phogrin-green fluorescent protein chimaera. Biochem J 1998;333:193–199. 17. Zahn TR, Macmorris MA, Dong W, Day R, Hutton JC. IDA-1, a Caenorhabditis elegans homolog of the diabetic autoantigens IA-2 and phogrin, is expressed in peptidergic neurons in the worm. J Comp Neurol 2001;429:127–143. 18. Lan MS, Lu J, Goto Y, Notkins AL. Molecular cloning and identification of a receptor-type protein tyrosine phosphatase, IA-2, from human insulinoma. DNA Cell Biol 1994;13:505–514. 19. Solimena M, Dirkx R, Hermel JM, Pleasic-Williams S, Shapiro JA, Caron L, Rabin DU. ICA512, an autoantigen of type I diabetes, is an intrinsic membrane protein of neurosecretory granules. EMBO J 1996;15:2102–2114. 20. Lobner K, Steinbrenner H, Roberts GA, Ling Z, Huang GC, Piquer S, Pipeleers DG, Seissler J, Christie MR. Different regulated expression of the tyrosine phosphatase-like proteins IA-2 and phogrin by glucose and insulin in pancreatic islets: relationship to development of insulin secretory responses in early life. Diabetes 2002;51:2982–2988. 21. Wasmeier C, Hutton JC. Secretagogue-dependent phosphorylation of phogrin, an insulin granule membrane protein tyrosine phosphatase homologue. Biochem J 1999;341:563–569. 22. Wasmeier C, Hutton JC. Secretagogue-dependent phosphorylation of the insulin granule membrane protein phogrin is mediated by cAMPdependent protein kinase. J Biol Chem 2001;276:31919–31928. 23. Ort T, Voronov S, Guo J, Zawalich K, Froehner SC, Zawalich W, Solimena M. Dephosphorylation of beta2-syntrophin and Ca2þ/ mu-calpain-mediated cleavage of ICA512 upon stimulation of insulinsecretion. EMBO J 2001;20:4013–4023. 24. Trajkovski M, Mziaut H, Altkruger A, Ouwendijk J, Knoch KP, Muller S, Solimena M. Nuclear translocation of an ICA512 cytosolic fragment couples granule exocytosis and insulin expression in {beta}-cells. J Cell Biol 2004;167:1063–1074. 25. Ort T, Maksimova E, Dirkx R, Kachinsky AM, Berghs S, Froehner SC, Solimena M. The receptor tyrosine phosphatase-like protein ICA512 binds the PDZ domains of beta2-syntrophin and nNOS in pancreatic beta-cells. Eur J Cell Biol 2000;79:621–630. 26. Gross S, Blanchetot C, Schepens J, Albet S, Lammers R, den Hertog J, Hendriks W. Multimerization of the protein-tyrosine phosphatase (PTP)-like insulin-dependent diabetes mellitus autoantigens IA-2 and IA-2beta with receptor PTPs (RPTPs). Inhibition of RPTPalpha enzymatic activity. J Biol Chem 2002;277:48139–48145. 27. Saeki K, Zhu M, Kubosaki A, Xie J, Lan MS, Notkins AL. Targeted disruption of the protein tyrosine phosphatase-like molecule IA-2
Traffic 2005; 6: 474–487
Recycling of Phogrin
28.
29.
30.
31.
32.
33. 34.
35.
36. 37.
38. 39.
40.
results in alterations in glucose tolerance tests and insulin secretion. Diabetes 2002;51:1842–1850. Kubosaki A, Gross S, Miura J, Saeki K, Zhu M, Nakamura S, Hendriks W, Notkins AL. Targeted disruption of the IA-2beta gene causes glucose intolerance and impairs insulin secretion but does not prevent the development of diabetes in NOD mice. Diabetes 2004;53:1684–1691. Cai T, Fukushige T, Notkins AL, Krause M. Insulinoma-associated protein IA-2, a vesicle transmembrane protein, genetically interacts with UNC-31/CAPS and Affects Neurosecretion in Caenorhabditis elegans. J Neurosci 2004;24:3115–3124. Leonard WJ, Depper JM, Crabtree GR, Rudikoff S, Pumphrey J, Robb RJ, Kronke M, Svetlik PB, Peffer NJ, Waldmann TA, et al. Molecular cloning and expression of cDNAs for the human interleukin-2 receptor. Nature 1984;311:626–631. Humphrey JS, Peters PJ, Yuan LC, Bonifacino JS. Localization of TGN38 to the trans-Golgi network: involvement of a cytoplasmic tyrosine-containing sequence. J Cell Biol 1993;120:1123–1135. Wasmeier C, Bright NA, Hutton JC. The lumenal domain of the integral membrane protein phogrin mediates targeting to secretory granules. Traffic 2002;3:654–665. Arvan P. Secretory protein trafficking: genetic and biochemical analysis. Cell Biochem Biophys 2004;40(Suppl. 3):169–178. Navone F, Di Gioia G, Jahn R, Browning M, Greengard P, De Camilli P. Microvesicles of the neurohypophysis are biochemically related to small synaptic vesicles of presynaptic nerve terminals. J Cell Biol 1989;109:3425–3433. Chen JW, Murphy TL, Willingham MC, Pastan I, August JT. Identification of two lysosomal membrane glycoproteins. J Cell Biol 1985;101:85–95. Ellis S, Mellor H. Regulation of endocytic traffic by rho family GTPases. Trends Cell Biol 2000;10:85–88. Fleming JC, Berger G, Guichard J, Cramer EM, Wagner DD. The transmembrane domain enhances granular targeting of P-selectin. Eur J Cell Biol 1998;75:331–343. Munro S. An investigation of the role of transmembrane domains in Golgi protein retention. EMBO J 1995;14:4695–4704. Marks MS, Ohno H, Kirchhausen T, Bonifacino JS. Protein sorting by tyrosine-based signals: adapting to the Ys and wherefores. Trends Cell Biol 1997;7:124–128. Bonifacino JS, Dell’Angelica EC. Molecular bases for the recognition of tyrosine-based sorting signals. J Cell Biol 1999;145:923–926.
Traffic 2005; 6: 474–487
41. Heilker R, Spiess M, Crottet P. Recognition of sorting signals by clathrin adaptors. Bioessays 1999;21:558–567. 42. Hurtley SM. Recycling of a secretory granule membrane protein after stimulated secretion. J Cell Sci 1993;106:649–655. 43. Patzak A, Winkler H. Exocytotic exposure and recycling of membrane antigens of chromaffin granules: ultrastructural evaluation after immunolabeling. J Cell Biol 1986;102:510–515. 44. Eliasson L, Proks P, Ammala C, Ashcroft FM, Bokvist K, Renstrom E, Rorsman P, Smith PA. Endocytosis of secretory granules in mouse pancreatic beta-cells evoked by transient elevation of cytosolic calcium. J Physiol 1996;493:755–767. 45. Tsuboi T, Rutter GA. Multiple forms of ‘kiss-and-run’ exocytosis revealed by evanescent wave microscopy. Curr Biol 2003;13:563–567. 46. Michael DJ, Geng X, Cawley NX, Loh YP, Rhodes CJ, Drain P, Chow RH. Fluorescent cargo proteins in pancreatic beta-cells: design determines secretion kinetics at exocytosis. Biophys J 2004;87:L03–L05. 47. Bauer RA, Overlease RL, Lieber JL, Angleson JK. Retention and stimulus-dependent recycling of dense core vesicle content in neuroendocrine cells. J Cell Sci 2004;117:2193–2202. 48. Futter CE, Connolly CN, Cutler DF, Hopkins CR. Newly synthesized transferrin receptors can be detected in the endosome before they appear on the cell surface. J Biol Chem 1995;270:10999–11003. 49. Yao J, Erickson JD, Hersh LB. Protein kinase A affects trafficking of the vesicular monoamine transporters in PC12 cells. Traffic 2004;5:1006–1016. 50. Vo YP, Hutton JC, Angleson JK. Recycling of the dense-core vesicle membrane protein phogrin in Min6 beta-cells. Biochem Biophys Res Commun 2004;324:1004–1010. 51. Wendler F, Page L, Urbe S, Tooze SA. Homotypic fusion of immature secretory granules during maturation requires syntaxin 6. Mol Biol Cell 2001;12:1699–1709. 52. Ang AL, Taguchi T, Francis S, Folsch H, Murrells LJ, Pypaert M, Warren G, Mellman I. Recycling endosomes can serve as intermediates during transport from the Golgi to the plasma membrane of MDCK cells. J Cell Biol 2004;167:531–543. 53. Owen DJ. Linking endocytic cargo to clathrin: structural and functional insights into coated vesicle formation. Biochem Soc Trans 2004;32: 1–14. 54. Bos K, Wraight C, Stanley KK. TGN38 is maintained in the trans-Golgi network by a tyrosine-containing motif in the cytoplasmic domain. EMBO J 1993;12:2219–2228.
487
