SEARCH

SEARCH BY CITATION

Keywords:

  • adenovirus;
  • cholera toxin;
  • endocytosis;
  • endoplasmic reticulum;
  • retrograde transport;
  • SNAP-tag;
  • SNAP-trap

Abstract

  1. Top of page
  2. Abstract
  3. Results
  4. Discussion
  5. Materials and Methods
  6. Acknowledgments
  7. References
  8. Supporting Information

Cholera toxin enters cells via an unusual pathway that involves trafficking through endosomes to the endoplasmic reticulum (ER). Whether the toxin induces its own pathway or travels along a physiological retrograde route is not known. To study its trafficking, we labeled cholera toxin B (CTB) or endogenous plasma membrane proteins with a small chemical compound, benzylguanine, which covalently reacts with the protein SNAP-tag. Using ER-targeted SNAP-tag as reporter, we found that transport of CTB to the ER depends on dynamin-2 and syntaxin 5. Plasma membrane proteins and a fluid-phase marker added to the medium were also transported to the ER. This flux was not affected by exposing cells to CTB but was inhibited by depleting syntaxin 5 and increased by depleting dynamin-2. As a control for confined intracellular localization of ER-targeted SNAP-tag we used adenovirus-5, which traffics to endosomes and then escapes into the cytosol. The virus did not react with ER-targeted SNAP but with cytosolic SNAP. Together, our results establish a new method (SNAP-trap) to study trafficking of different cargo to the ER and the cytosol and provide evidence for the existence of a constitutive pathway from the cell surface to the ER.

The endoplasmic reticulum (ER) is the point of departure for newly synthesized proteins embarking into the secretory pathway. After leaving the ER through specialized exit sites, soluble proteins traffic through the Golgi complex and are either directed to endosomes or secreted to the extracellular space. Similarly, membrane proteins are targeted to the plasma membrane (PM) and other membranes. The forward-directed membrane flow is compensated by retrograde transport of lipids and proteins to maintain organelle integrity [1-4]. For example, ER-resident proteins that escape are continuously shuttled back from the cis-Golgi to the ER [5]; and mannose-6 phosphate receptors are recycled back to the trans-Golgi network (TGN) after conveying cargo proteins to endosomes [6, 7].

Certain toxins and viruses traffic from the outside of the cell all the way to the ER where they penetrate the membrane to intoxicate or infect the cell [8]. The pathway taken by cholera toxin (CT) involves binding to the ganglioside GM1 at the PM, internalization and trafficking through early endosomes and the TGN to the ER [9]. Like CT, simian virus 40 (SV40) binds to GM1 but after passing early endosomes, the virus traffics through late endosomes to the ER [10-12]. It is possible that these pathogens exploit the host cell's membrane-trafficking system and rely on constitutive retrograde pathways from the PM to the ER. However, the physiological role of such a pathway is not known as only few endogenous ligands that are targeted to the ER are known i.e. autocrine motility factor [13]. The majority of viruses including human adenovirus type 5 (hAdV-5) enter cells via endosomes from which they directly escape into the cytosol [11].

The movement of fluorescently labeled viruses and toxins through the cell can be tracked by fluorescence microscopy [reviewed in [14]]. While colocalization or cotracking with e.g. endosomal markers has been successfully applied to study early endocytic transport steps [11, 12, 15, 16], it is difficult to apply this technique to detect small pathogens in the ER or the cytosol. Markers of these compartments are too ubiquitously dispersed to allow colocalization studies without the likelihood of random colocalization. Trafficking to the ER and the cytosol has mainly been studied by time-consuming techniques such as electron microscopy [17-19] or by engineering pathogens with reporter peptides that react with enzymes only present in these compartments. For example, acceptor peptides for N-linked glycosylation were used as reporters for ER trafficking [20, 21].

We developed a simple assay to examine the trafficking of labeled particles or proteins from the extracellular space or PM to the ER and the cytosol [22]. Our method is based on the reporter protein SNAP-tag that covalently and irreversibly reacts with a small chemical compound, benzylguanine (BG) [23]. Viruses, toxins, membrane proteins or receptor ligands are labeled with an amine-reactive BG derivative to transform them into SNAP-tag substrates. Cells are transfected with constructs targeting SNAP-tag to either the cytosol or the ER to capture trafficking particles. The reaction between SNAP and BG-labeled proteins is monitored using a gel mobility shift assay. Using this method (dubbed SNAP-trap), we found that trafficking of cholera toxin B (CTB corresponding to the homopentameric CT devoid of the toxic A subunit) to the ER depends on dynamin-2 and syntaxin 5. Furthermore, our results showed that proteins labeled with BG on the PM and a fluid-phase marker trafficked to the ER suggesting the existence of a physiological pathway from the cell surface to the ER.

Results

  1. Top of page
  2. Abstract
  3. Results
  4. Discussion
  5. Materials and Methods
  6. Acknowledgments
  7. References
  8. Supporting Information

CTB is captured by ER-localized SNAP-tag

For labeling, CTB was mixed with BG at a molar ratio of 1:10. Mass spectrometry analysis of the labeled CTB indicated that up to nine BG molecules were attached to each subunit of the toxin (Figure S1 and Appendix S1). Then, BG-CTB was incubated with purified SNAP-tag (Figure 1A) for 30 min, and samples analyzed by SDS–PAGE. Apparently, SNAP-tag formed two covalent linkage products with BG-labeled CTB as two bands appeared on the gel that were absent when SNAP-tag and CTB were analyzed alone (Figure 1B, black arrows). The faster migrating band corresponded to a linkage product between SNAP (20 kDa) and a subunit (13 kDa) of BG-CTB. The slower migrating band was assigned as CTB linked to two SNAP-tag proteins.

image

Figure 1. BG-labeled CTB reacts with SNAP-tag in vitro. A) Schematic representation of the reaction between SNAP-tag and BG bound to CTB. B) CTB, labeled with BG (BG-CTB) was incubated for 30 min at 37°C with purified SNAP-tag and analyzed by SDS–PAGE followed by Coomassie Blue staining. Two reaction products were formed (black arrows). As controls, SNAP-tag and BG-CTB were analyzed.

Download figure to PowerPoint

Next, CV-1 cells were transfected with a construct that targets myc-tagged SNAP-tag to the ER and confines it there (ER-SNAP, Figure 2A). It has an ER signal sequence in the N-terminus and a myc-tag as well as a KDEL retention sequence at the C-terminus [22]. The cells were treated with a membrane permeable, fluorescent SNAP-tag substrate, TMR-Star. After fixation, they were stained with antibodies against the ER marker calnexin and analyzed by confocal microscopy. As shown in Figure 2B, the fluorescence from TMR-Star overlapped with the calnexin staining indicating that ER-SNAP was localized in the ER and enzymatically active.

image

Figure 2. BG-labeled CTB reacts with ER-targeted SNAP in cells. A) Schematic representation of the construct used to target SNAP-tag to the ER (ER-SNAP). The N-terminus of SNAP-tag is preceded by an ER signal sequence (ss). A myc-epitope (myc) followed by an ER retention signal sequence (KDEL) are fused to the C-terminus of SNAP-tag. B) CV-1 cells were transiently transfected with ER-SNAP. After 24 h, cells were treated with TMR-Star, fixed and stained with antibodies against calnexin. Images were taken with a confocal fluorescence microscope. The scale bar represents 10 µm. C) CV-1 cells transiently transfected with ER-SNAP were exposed to BG-CTB for the indicated times (0.5 µg BG-CTB in 1 mL medium per 250 000 cells). Where indicated, 10 µg/mL BFA was added to the cell medium 5 min prior to the addition of BG-CTB. Cells were lysed and analyzed by SDS–PAGE followed by immunoblot analysis using an antibody against the myc-epitope of ER-SNAP. Four reaction products between ER-SNAP and BG-CTB were formed. D) Same as in (C) but cells were exposed 6 h to BG-CTB. Where indicated 50 µm CI-976 was added to the cell medium 5 min prior to the addition of BG-CTB. E) Same as in (C) but cells were exposed 12 h to BG-CTB. Where indicated, pharmacological inhibitors were added to the cell medium 5 min prior to the addition of BG-CTB using the following concentrations: BFA, 10 µg/mL; Rapamycin, 15 nM; Dynasore, 80 µm; Cytochalasin D, 10 µg/mL; Nocodazole, 5 µg/mL and Bafilomycin A1 100 nm. A nonspecific band is marked with an asterix.

Download figure to PowerPoint

ER-SNAP expressing cells were then exposed to BG-labeled CTB. Under these conditions, BG-CTB must traffic to the ER to be captured by ER-SNAP. After different times, cells were lysed with a denaturing buffer to avoid post-lysis reactions. Samples were analyzed by SDS–PAGE followed by immunoblot analysis using an antibody against the myc-epitope of ER-SNAP. In addition to unreacted ER-SNAP running at a molecular size of 20 kDa, four reaction products between SNAP and BG-CTB were detected 3 h after exposure of cells to the toxin (Figure 2C). The reaction products migrated in a range between 50 and 150 kDa indicating that multiple ER-SNAP reacted with BG-CTB. The intensity of the bands increased over time and after 9 h a fifth reaction product became visible on the western blot (red arrow). It is noteworthy that crosslinking in the ER was more efficient than the in vitro crosslinking depicted in Figure 1. This was most likely due to the increased concentration of both reaction partners when they are compartmentalized in the ER, and by the longer incubation time. The products were not formed when the same experiment was performed in the presence of brefeldin A (BFA). BFA disrupts the Golgi and blocks CT intoxication [24]. Thus, BG-CTB was only captured by ER-SNAP if the entry route of CT was intact.

To test whether the reaction between BG-CTB and ER-targeted SNAP-tag occurred within the ER and not in secretory compartments beyond the ER, ER-SNAP expressing cells were treated with an inhibitor of ER exit site formation, CI-976 [25]. We have shown earlier that CI-976 inhibits secretion of a soluble substrate [26] and ER-export of a membrane associated protein [27]. When cells treated with 50 µm CI-976 were exposed to BG-CTB for 6 h, the toxin was captured by ER-SNAP with a similar efficiency as in control cells (Figure 2D). As shown in Figure S2, 50 µm CI-976 blocked ER-export of caveolin-1, a membrane-associated substrate. These results indicate that SNAP-trapping occurred predominantly, if not exclusively, within the lumen of the ER.

Effect of pharmacological inhibitors on CTB trafficking

To characterize the pathway of CTB to the ER, we tested several pharmacological inhibitors. Rapamycin, an inhibitor of mTOR only slightly affected the trafficking of CTB to the ER (Figure 2E). Similar results were obtained with nocodazole, a drug that interferes with polymerization of microtubules. In contrast, dynasore (an inhibitor of dynamin-2) and cytochalasin D (an inhibitor of actin filament assembly), largely blocked SNAP-trapping of BG-CTB. A moderate decrease was caused by bafilomycin A1, an inhibitor of the vacuolar ATPases responsible for the acidification of endosomes, lysosomes and the TGN. Thus, ER-trafficking of CTB, as judged by SNAP-trap, was dependent on dynamin, actin filaments and to a lesser extent on the acidification of vacuolar organelles.

Trafficking of CTB to the ER depends on dynamin-2 and syntaxin 5

Using the SNAP-trap assay, we tested the involvement of dynamin-2 and syntaxin 5 in trafficking of CTB to the ER. Dynamin-2 supports uptake [28] and syntaxin 5 retrograde endosome to TGN transport of CT [29]. Dynamin-2 and syntaxin 5 were silenced using small interfering RNAs (siRNAs). The performance of the siRNAs was tested by immunoblot analysis of CV-1 cell lysates with protein-specific antibodies. Both siRNAs markedly reduced the expression level of their target protein compared to nontargeting control siRNA (Figure 3A,B).

image

Figure 3. Dynamin-2 is required for the internalization of CTB, syntaxin 5 for intracellular trafficking to the ER. A) CV-1 cells were transfected with a nontargeting siRNA (negative control) or with a siRNA against dynamin-2. Twenty microgram of cell lysate was analyzed by SDS–PAGE followed by immunoblot analysis with an antibody against dynamin-2 or tubulin. B) Same as in (A) but syntaxin 5 was silenced. C) CV-1 cells treated with the indicated siRNAs were exposed to BG-CTB for 12 h. Cell lysates were analyzed for the formation of reaction products between ER-SNAP and BG-CTB as in Figure 2E. The depicted images are parts of the same immunoblot shown in Figure S3A. D) Quantification of the band intensities corresponding to reaction products between BG-CTB and ER-SNAP. Data points are relative to the negative control (nontargeting siRNA). Bars represent the means ± SEM of three independent experiments. E) CV-1 cells treated with the indicated-siRNAs were exposed to Biotin-SS-CTB (0.5 µg BG-CTB in 1 mL medium per 250 000 cells) at 4°C for 2 h. Then, cells were shifted to 37°C and incubated for the indicated times. Cells were either treated with TCEP for 10 min or left untreated. After fixation, cells were stained with an antibody against biotin and analyzed by flow cytometry. F) CV-1 cells were treated with the indicated-siRNA. Biotin-SS-CTB was allowed to internalize at 37°C for 30 min. The amount of internalized CTB was calculated as the ratio of the mean fluorescence intensity of TCEP-treated and untreated samples. Data points are relative to negative controls (nontargeting siRNA). Bars represent the mean ± SEM of three independent experiments.

Download figure to PowerPoint

Cells treated with the siRNAs for 48 h were subjected to a second transfection with a plasmid encoding ER-SNAP. After 24 h, they were exposed to BG-CTB and lysed after a further 12 h. As shown in Figure 3C,D, SNAP-trapping of the toxin was significantly reduced confirming that ER-trafficking of CTB depended on these proteins [28, 29].

To test whether dynamin-2 and syntaxin 5 were required for endocytosis or subsequent steps in the pathway, we used CTB labeled with thiol-cleavable biotin (Biotin-SS-CTB). It was bound in the cold to cells treated with siRNAs. The cells were then incubated either with tris(2-carboxyethyl)phosphine (TCEP), a membrane-impermeable reducing agent that removes the biotin from surface accessible Biotin-SS-CTB or with buffer as a control. After fixation, cells were stained with an antibody against biotin and analyzed by fluorescence-activated cell sorting (FACS). In cells transfected with a control siRNA and held in the cold to avoid endocytosis, the fluorescence decreased dramatically after the TCEP treatment (Figure 3E). In contrast, after 30 min at 37°C, the majority of the biotin was not removed by TCEP, indicating that Biotin-SS-CTB had been internalized. Similar results were obtained with cells in which syntaxin 5 was silenced (Figure 3F). However, in cells transfected with a siRNA targeting dynamin-2, internalization of Biotin-SS-CTB was reduced by 40%. The moderate reduction was consistent with results from a previous study indicating that the toxin enters cells via multiple routes only some of which depend on dynamin-2 [30]. Syntaxin 5 was evidently involved in postinternalization steps during CTB trafficking and dynamin-2 in the endocytic uptake.

Trafficking of cell-surface proteins to the ER

To test whether transport from the PM to the ER exists as a constitutive pathway in the cells, we labeled cell-surface proteins with BG in the cold for 30 min. Cells were washed and lysed with a mild detergent (NP40), the lysate incubated with purified and fluorescently labeled 488-SNAP and the proteins analyzed by SDS–PAGE followed by fluorescence gel scanning. Figure 4A indicates that BG-labeling and conjugation to SNAP-tag occurred as several high-molecular weight adducts were seen in the sample containing BG-labeled membrane proteins.

image

Figure 4. BG-labeled PM proteins traffic to the ER. A) CV-1 cells were labeled with BG on ice for 30 min or left untreated. Cells were lysed with NP40 and postnuclear supernatants containing solubilized membrane proteins (MPs) were incubated with Alexa Fluor 488 labeled SNAP (488-SNAP). Samples were analyzed by SDS–PAGE followed by fluorescent gel scanning. B) CV-1 cells were either labeled with BG or left untreated. Then, SV40 was added to cells or as a control no virus was added. After 24 h, cells were fixed, stained with antibodies against the viral T-antigen and analyzed by flow cytometry. C) CV-1 cells expressing ER-SNAP were labeled with BG on ice. Cells were then shifted to 37°C and incubated for the indicated times. Cells were lysed and analyzed by SDS–PAGE followed by immunoblot analysis with antibodies against the myc-epitope of ER-SNAP. D) CV-1 cells expressing the indicated SNAP variants were labeled with BG where indicated. Cells were incubated for 8 h at 37°C and then analyzed. BFA, CTB or SV40 were added to the cell medium where indicated. E) Same as in (D) but where indicated, pharmacological inhibitors were added to the cell medium or cells were treated with siRNAs. F) CV-1 cells expressing ER-SNAP were exposed to BG-labeled SV40. After 24 h cells were lysed and analyzed by SDS–PAGE followed by immunoblot analysis with an antibody against the myc-epitope of ER-SNAP. Where indicated, bafilomycin A1 was added to the cell medium or cells were treated with a siRNA against dynamin-2. G) BG-Atto532 was added to the medium of CV-1 cells expressing ER-SNAP. Where indicated, pharmacological inhibitors were added to the cell culture medium or cells were treated with siRNAs. After the indicated incubation times, cells were lysed and analyzed by fluorescent gel scanning. Gels were subsequently blotted onto a PVDF membrane and analyzed with an antibody against the myc-epitope of ER-SNAP. H) Same as in (G), but where indicated SV40 or CTB was added to the cell culture medium.

Download figure to PowerPoint

When BG-labeled cells were infected with SV40 at a low multiplicity of infection for 24 h, the amount of infected cells (SV40 T-antigen positive) was comparable to controls in which nonlabeled cells were infected (Figure 4B). This suggested that membrane trafficking in BG-labeled cells was largely intact.

Next, cells expressing ER-SNAP were labeled with BG in the cold, carefully washed and then shifted to 37°C. After different time periods, cells were lysed with a SDS-containing buffer and analyzed by western blot. Proteins captured by ER-SNAP started to appear after 2 h and became increasingly prominent up to 8 h with a half time of roughly 4 h (Figures 4C and S3B). Thus the time course was similar to that measured for the arrival of CT (Figure 2C) [20] and SV40 [18] in the ER.

Conjugates between the labeled PM proteins and ER-SNAP were not formed if cells were treated with BFA (Figures 4D and S3C). Nor did they form when the cytosolic SNAP-tag was expressed instead of ER-SNAP. High concentrations of CTB or SV40 did not affect the amount PM proteins captured by ER-SNAP, indicating that the pathogens did not stimulate or inhibit the retrograde pathway. While the flux of BG-labeled proteins to the ER was only marginally affected by nocodazole, it was inhibited by cytochalasin D and elevated by bafilomycin A1 (Figure 4E). Furthermore, depletion of syntaxin 5 inhibited the transport while depletion of dynamin-2 increased it.

These observations indicated that the pathway taken was not identical to that used by CTB or SV40. We could confirm in this system that treatment of cells with bafilomycin A1 or siRNAs against dynamin-2 inhibited transport of SV40 to the ER (Figure 4F) [12]. Although the perturbation analysis implied distinct pathways and cellular mechanisms, it was apparent that the SNAP-conjugation was not the result of nonspecific redistribution of ER-SNAP into other compartments.

Fluid transport to the ER

We next examined whether components in the extracellular fluid were transported to the ER. For this, we used a soluble, membrane-impermeable fluorophore coupled to BG (Atto532-BG). Cells expressing ER-SNAP were incubated with this compound, lysed and analyzed by SDS–PAGE and fluorescence gel scanning. After 1–2 h incubation almost no Atto532-BG reacted with ER-SNAP (Figures 4G and S3D). However, after prolonged incubation, ER-SNAP was labeled by the fluid-phase marker as a fluorescent band appeared on the gel. Like trafficking of PM proteins to the ER, fluid transport was inhibited by BFA, cytochalasin D and by silencing of syntaxin 5 (Figure 4G). Moreover, silencing of dynamin-2 also increased the fluid stream while addition of CTB and SV40 had no effect (Figure 4H). Trafficking of Atto532-BG thus seemed to follow the same pathway as the PM proteins.

Human adenovirus type 5 (hAdV-5) is captured by cytosolic SNAP

To test whether ER-targeted SNAP-tag was present in compartments other than the ER, we tested hAdV-5, which enters early endosomes and escapes into the cytosol by rupturing the endosomal membrane [31] (Figure 5A). Thus, hAdV-5 was not expected to react with ER-targeted SNAP.

image

Figure 5. BG-labeled adenovirus 5 reacts with cytosol-SNAP but not with ER-SNAP. A) Schematic representation of the endocytosis pathways of CTB and hAdV-5. Both traffic to early endosomes. hAdV-5 escapes from early endosomes to the cytosol whereas CTB traffics from early endosomes to the trans-Golgi and subsequently to the ER. BG-labeled CTB reacts with ER-SNAP, BG-labeled hAdV-5 with cytosol-SNAP. B) 250 000 A549 cells expressing the indicated SNAP variants (ER-SNAP or cytosol-SNAP) were infected with 6.5 µg BG-labeled or unlabeled hAdV-5 or were left untreated. After 7 h postinfection, cells were lysed and analyzed by SDS–PAGE followed by immunoblot analysis with an antibody against the myc-epitope of SNAP or with an antibody against the hexon protein of hAdV-5.

Download figure to PowerPoint

A549 cells transfected with either ER-SNAP or a construct targeting myc-tagged SNAP to the cytosol (Cytosol-SNAP) were exposed to BG-labeled hAdV-5 for 7 h. After cell lysis, reaction products between SNAP and hAdV-5 proteins were analyzed by western blot. While unlabeled hAdV-5 did not react with SNAP, BG-labeled hAdV-5 efficiently reacted with cytosol-SNAP but barely with ER-SNAP (Figures 5B and S2E). This confirmed that ER-targeted SNAP was almost completely absent from the endosomes and the cytosol.

Discussion

  1. Top of page
  2. Abstract
  3. Results
  4. Discussion
  5. Materials and Methods
  6. Acknowledgments
  7. References
  8. Supporting Information

Using SNAP-trap, we were able to monitor the arrival of different endocytozed proteins and reagents in the ER and the cytosol. Endogenous enzymes confined to these compartments have, previously, been used as reporters for protein trafficking. For example, the ER-resident oligosaccharyltransferase [20, 21] Golgi-resident sialyl- [32] and sulfotransferase [33, 34] have been used. Often these strategies require the addition of an acceptor-peptide for cargo proteins to be recognized and modified by the enzymes. A major advantage of SNAP-trap is that cargo proteins are labeled with a small lysine-reactive compound without the need of engineering the amino acid sequence. This minimizes changes in the tertiary and quaternary structures of the proteins and allows the study of genetically unmodified cargo e.g. wild-type virus strains.

When provided with specific signal sequences and localization tags, SNAP-tag can be targeted to diverse cell compartments such as the Golgi, mitochondria and nucleus [23, 35-37]. In addition to ER-SNAP, we used cytosol-targeted SNAP-tag to follow the escape of BG-labeled adenovirus from endosomes. The strong signal observed was consistent with previous electron microscopy studies showing that adenoviruses efficiently escape from endosomes [17]. Using cytosolic SNAP-tag, we have previously shown that SV40 penetrates from the ER into the cytosol, albeit with significantly lower efficiency [22]. Very recently, SNAP-tag targeted to the Golgi was used to identify PM proteins that traffic to the Golgi [37]. Such SNAP-trap assays are especially useful when analyzing how incoming nonenveloped viruses and toxins cross the lipid bilayer. In the future, SNAP-trap might also become useful in assays used for the development of delivery methods in gene therapy and pharmacology.

Trafficking to the ER is a mandatory step in CT intoxication because the catalytic A-subunit requires ER-localized thiol oxidoreductases and dislocation factors to access the cytosol [9, 38, 39]. Our results showed that to reach the ER, CTB depended on dynamin-2 and syntaxin 5. The results were consistent with previous studies showing that the toxin is partially dependent on dynamin-2 for endocytosis and that it requires syntaxin 5, a SNARE protein, for intoxication [28, 29]. The syntaxin 5 dependence indicated that CT passes early endosomes and the TGN [29, 40]. From the TGN, it may be transferred directly to the ER without moving through the Golgi stacks [41].

Diversion of cargo can also occur from organelles deeper in the endocytic pathway. Cation-independent mannose 6-phosphate receptor is thought to move from late endosomes to the TGN [42, 43] and free cholesterol directly from lysosomes to the ER [44]. Viruses that use gangliosides as receptors such as SV40 and mouse polyoma virus are likely to pass from maturing late endosomes or endolysosomes to the ER [12, 15]. The infectious entry of SV40 also depends on syntaxin 5, and the endosome to ER transport step requires dynamin-2 [12]. However unlike CT, the intracellular transport steps are inhibited by bafilomycin A1 and nocodazole. The pathway taken by CT is therefore most likely distinct.

A common feature of toxins and viruses that traffic to the ER is their capacity to induce membrane curvature [45, 46]. It is thus possible that they actively induce certain trafficking steps and do not reach the ER through a physiological pathway. Using an alternative strategy to investigate membrane trafficking from the PM to the ER that does not rely on pathogens or toxins, we found a fraction of BG-labeled PM proteins that reacted with ER-SNAP under unperturbed conditions.

Transport pathways from the extracellular space to the ER have previously been observed for the lipid-raft-dependent internalization of autocrine motility factor in tumor cells [13, 47] and for antigen uptake in antigen-presenting cells. In the latter case, peptides originating from endocytozed antigens are loaded onto newly synthesized major histocompatibility class I proteins in the ER. In some cases, phagosomes were found to interact with the ER [48-50], suggesting a direct delivery route of endocytozed antigens to the ER. Moreover, soluble proteins exogenously added to cells were found to accumulate in the ER [51].

Materials and Methods

  1. Top of page
  2. Abstract
  3. Results
  4. Discussion
  5. Materials and Methods
  6. Acknowledgments
  7. References
  8. Supporting Information

Materials

Recombinant CTB was from Crucell (SBL vaccines). Anti-syntaxin 5A antibody (clone 5A6) and anti-myc antibody (clone 9E10) were from Sigma; anti-dynamin-2 antibody was from BD Transduction Laboratories. Anti-biotin antibody was from Bethyl Laboratories. Anti-Hexon 24240 antibody was from Abcam. Polyclonal anti-calnexin antibody was produced in-house. BG-GLA-NHS, SNAP-Cell TMR-Star and SNAP-Surface 532 were from New England Biolabs. Alexa Fluor 488 succinimidyl ester was from Molecular Probes. Sulfo-NHS-SS-Biotin was from Pierce. Rapamycin, nocodazole, bafilomycin A1, BFA, dynasore and cytochalasin D were from Sigma. Lipofectamine RNAiMAX from Invitrogen. All siRNAs were from Qiagen.

Cell Culture, viruses and plasmids

All cells were from ATCC. CV-1 cells were grown in DMEM (Gibco) supplemented with 10% FCS and 1% Glutamax. A549 cells were grown in DMEM supplemented with 7.5% FBS and 1% nonessential amino acids. SV40 was grown in CV-1 cells and purified as described in [52]. Adenovirus 5 was grown and purified as described in [53]. ER-SNAP and cytosol-SNAP plasmids were described in [22].

siRNA-mediated gene silencing

The siRNAs targeting dynamin-2 (SI02654687) and syntaxin 5 (SI00048636) were transfected into CV-1 cells at 20 nm using lipofectamine RNAiMAX (Invitrogen) according to the manufacturer's recommendations. Nontargeting siRNA (AllStarsNeg; QIAGEN) was used as control siRNA. Cells were analyzed after 72 h. For SNAP experiments, cells were trypsinized 48 h after siRNA transfection and SNAP plasmids were transfected using AMAXA.

Immunofluorescence imaging

Cells grown on coverslips were fixed using 4% formaldehyde in PBS. Cells were permeabilized using 0.05% saponin and 1% BSA in PBS and incubated with the appropriate primary and secondary antibodies. Coverslips were mounted on slides using Immumount (Thermo Fisher Scientific). Imaging was performed on an inverted confocal microscope system (LSM 510 Meta; Carl Zeiss, Inc.) using a 100× 1.4 NA objective.

SV40 infection assay

Cells were infected and analyzed by flow cytometry as described in [22].

SNAP experiments

CV-1 cells or A594 cells were trypsinized and 5 µg SNAP plasmid was transfected into 106 cells using AMAXA (for CV-1 cells) and Neon (for A594 cells). After transfection, cells from different transfections were mixed and plated into 12-well plates. After 24 h, cells were exposed to BG-labeled CT, SV40 or hAdV-5. After appropriate incubation times, cells were washed twice with PBS and directly lysed with 100 μL SDS–PAGE sample buffer and analyzed by SDS–PAGE and western blot; 20 μL cell lysate was loaded on a gel.

Labeling of CTB with BG

Recombinant CTB (stock solution 10 mg/mL) was diluted with a fresh solution of NaHCO3 (20 mg/mL, pH ≈ 8) to a final concentration of 100 µm. Then, BG-GLA-NHS [stock solution 10 mg/mL in water-free dimethyl sulphoxide (DMSO)] was added to a final concentration of 1 mm (Labeling ratio 1:10). The reaction mixture was mixed well and rotated for 1 h at room temperature and quenched with 100 mm Tris HCl pH 8 for 15 min. The solution was washed four times with three volumes of PBS and after each wash the volume was reduced using a centrifugal filter device with a cut off of 5000 NMWL.

Labeling of hAdV-5 with BG

Double CsCl purified hAdV-5 was labeled with BG-GLA-NHS at a molar ratio of free BG-GLA-NHS to viral protein of 10/1 in 400 μL 0.1 M NaHCO3, 50 mm NaCl, 1 mm MgCl2 pH 8.2 for 1 h and quenched with 100 mm Tris HCl pH 8. Labeled hAdV-5 was repurified by CsCl density-gradient centrifugation similar to fluorophore labeling of hAdV5 [54].

Labeling of surface proteins with BG-GLA-NHS

CV-1 cells transfected with SNAP plasmids were washed twice with PBS and labeled for 30 min on ice with a fresh solution of 3.3 µg/mL BG-GLA-NHS in PBS. Cells were washed three times with complete medium.

Fluid-phase transport to the ER using Atto532-BG

CV-1 cells transfected with 5 µg of ER-SNAP plasmid were plated into 12-well plates. Cells were washed three times with complete medium and then incubated with 5 µm SNAP-surface-532 in complete medium. After various incubation times, cells were washed three times with complete medium and incubated for a further 30 min. Cells were lysed, boiled at 65°C and analyzed by SDS–PAGE followed by fluorescent gel scanning (Typhoon FLA 9000, Cy3 channel laser 532).

CT internalization assay

CT labeled with Sulfo-NHS-SS-Biotin was bound to CV-1 cells in the cold for 2 h. Cells were either directly treated with a TCEP solution (50 mm pH 7.4 in PBS, 10 min at 37°C) or after 30 min incubation at 37°C. Subsequently, cells were fixed, stained with an antibody against biotin and analyzed by FACS. Data were analyzed with FlowJo.

Labeling of SNAP-His with Alexa Fluor 488

SNAP-His (32.5 µg) in 100 μL NaHCO3 solution (20 mg/mL, pH ≈ 8) was labeled with Alexa Fluor 488 succinimidyl ester in a molar ratio of 1:10 for 1 h at room temperature. After quenching with 100 mm Tris HCl pH 8 for 15 min the solution was washed four times with 300 μL PBS and after each wash the volume was reduced using a centrifugal filter device with a cut off of 5000 NMWL.

Preparation of solubilized cell membranes

CV-1 cells were grown in a 10 cm dish and labeled on ice with a fresh solution of 3.3 µg/mL BG-GLA-NHS in PBS for 30 min. Cells were then washed two times with complete medium and two times with PBS. Cells were scraped in PBS, centrifuged at 300× g for 5 min at 4°C and resuspended in 500 μL lysis buffer (0.5% NP40, 150 mm NaCl, 20 mm Hepes pH 7.4 and protease inhibitors). After cell lysis (20 min at 4°C) nuclei were centrifuged (1100× g for 10 min). Alexa Fluor-488 (10 µg) labeled SNAP was added to the sample and the reaction mixture was incubated at 37°C for 30 min. Then, SDS sample buffer was added and samples were analyzed by SDS–PAGE followed by fluorescent gel scanning (Typhoon FLA 9000, Cy2 channel laser 532).

Acknowledgments

  1. Top of page
  2. Abstract
  3. Results
  4. Discussion
  5. Materials and Methods
  6. Acknowledgments
  7. References
  8. Supporting Information

We thank members of the Helenius laboratory for discussions. This work was supported by an ERC advanced investigator grant to AH, by a lipidX grant to AH and UFG and by a SNF grant to UFG.

References

  1. Top of page
  2. Abstract
  3. Results
  4. Discussion
  5. Materials and Methods
  6. Acknowledgments
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Results
  4. Discussion
  5. Materials and Methods
  6. Acknowledgments
  7. References
  8. Supporting Information
FilenameFormatSizeDescription
tra12018-sup-0001-FigureS1.docWord document48KFigure S1: Mass spectrometry experiment to determine the number of installed BG molecules on CTB. Mass spectra of CTB labeled with BG at a molar ratio of 1:10 (upper panel) and of unlabeled CTB (lower panel).
tra12018-sup-0002-FigureS2.docWord document101KFigure S2: Cycloheximide (CHX) washout experiment to test the functionality of the drug CI-976. This assay is described in detail in [27]. Cells were transfected with HA-tagged caveolin-1 and cultured in medium containing 0.5 mm CHX to inhibit protein synthesis. After 3 h, when cells were attached to cover slips, CHX was washed away and cells were either incubated in medium containing 50 µm CI-976 or in medium without a drug. After a further 45 min, cells were fixed, stained with antibodies against HA and giantin and analyzed by confocal microscopy. Scale bars represent 10 µm.
tra12018-sup-0003-FigureS3.docWord document237KFigure S3: Full scans of the western blots and gels. A part of (A) is depicted in Figure 2E. A part of (B) is shown in Figure 4C, a part of (D) is shown in Figure 4G and a part of (E) is shown in Figure 5B. Unspecific bands are marked with an asterix throughout.
tra12018-sup-0004-Appendix S1.docxWord 2007 document22K Appendix S1: Considerations when using the SNAP-trap method.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.