4062 Research Article

IntroductionNormal cellular prion protein (PrPc) is aglycosylphosphatidylinositol (GPI)-anchored glycoprotein that isexpressed predominantly in the brain, as well as in the spinal cordand at lower levels in leukocytes and some peripheral tissues (Paulyand Harris, 1998; Prusiner, 1998; Weissmann and Flechsig, 2003).Prion diseases are caused by the conversion of PrPc into a protease-resistant misfolded isoform (PrPsc). The prion diseases includetransmissible spongiform encephalopathies such as scrapie in sheepand goats, bovine spongiform encephalopathy, and Creutzfeldt-Jakob disease in humans (Prusiner, 1998). Formation of PrPsc inall of these disorders occurs when nerve cells are exposed to PrPsc,which then converts PrPc to aggregated deposits of PrPsc. In scrapie,it is not yet clear whether a loss of function of PrPc or a gain offunction by PrPsc is responsible for the neuronal loss and spongiformchanges in the brain. However, the observation that clinicalsymptoms occur without any obvious scrapie deposits (Collinge etal., 1990; Medori et al., 1992) suggests that it is the loss of normalPrPc function, not the formation of PrPsc deposits, that causes priondisease (Aguzzi et al., 1997).

To understand the pathogenesis of scrapie, it is important to knowthe mode of PrPc internalization in the cell. The biological role ofPrPc internalization is unknown, although it is stimulated by Cu2+

(Pauly and Harris, 1998). Like many GPI-anchored proteins, PrPc

is found clustered in the sphingomyelin- and cholesterol-richdomains of the membrane known as lipid rafts (Taylor and Hooper,2006). Although PrPc is present in rafts, it has been widely acceptedthat, at least in neuronal cells, PrPc is internalized via clathrin-coatedpits. Support for this view has come from multiple studies usingdifferent cell models. The clathrin-dependent pathway for PrPc

internalization was first proposed by the Harris laboratory using amouse N2a cell line stably transfected with chicken PrPc (Shynget al., 1994). Their evidence that PrPc was internalized via clathrin-coated pits came from immunogold labeling of PrPc in clathrin-coated pits, from blocking PrPc internalization by hypertonicsucrose, which disrupts clathrin lattices, and from the detection ofPrPc in clathrin-coated-vesicle preparations from chicken brain.

Additional evidence that PrPc is internalized via clathrin-coatedpits has come from kinetic studies using primary neurons. In thesestudies, the Morris laboratory found that the rate of PrPc

internalization was similar to that of transferrin, a cargo that isknown to be internalized by clathrin-mediated endocytosis (Sunyachet al., 2003). They also found, by immunogold labeling, thatendogenous PrPc was localized to clathrin-coated pits in N2a cells.Studies, primarily from the Hooper laboratory, using the humanneuronal cell line SHY5Y stably expressing mouse PrPc have alsosupported the clathrin-dependent internalization of PrPc in thepresence of Cu2+ (Taylor et al., 2005). This group found that theCu2+-stimulated internalization of PrPc was blocked by the drugtyrphostin A23, which also blocks internalization of the transferrinreceptor. More recently, the Taylor laboratory showed that depletingcells of the LRP1 receptor, a scavenger receptor that is internalizedvia clathrin-mediated endocytosis, blocks Cu2+-stimulatedinternalization of PrPc (Taylor and Hooper, 2007). However, it isnot clear that this effect of LRP1 knockdown is due to a decreasein receptor internalization because LRP1 has also been shown toaffect the biosynthetic trafficking of PrPc (Parkyn et al., 2008).

Although many studies have indicated that PrPc is internalizedvia clathrin-mediated endocytosis, several studies have suggestedthat PrPc is also internalized via cholesterol-enriched raft and/or

To understand the role of clathrin-mediated endocytosis in theinternalization of normal cellular prion protein (PrPc) inneuronal cells, N2a cells were depleted of clathrin by RNAinterference. PrPc internalization via the constitutive endocyticpathway in the absence of Cu2+ and the stimulated pathway inthe presence of Cu2+ were measured in both control andclathrin-depleted cells. Depletion of clathrin had almost no effecton the internalization of PrPc either in the presence or absenceof Cu2+, in contrast to the marked reduction observed intransferrin uptake. By contrast, the internalization of PrPc wasinhibited by the raft-disrupting drugs filipin and nystatin, andby the dominant-negative dynamin-1 mutant dynamin-1 K44A,

both in the presence and absence of Cu2+. The internalized PrPc

was found to colocalize with cargo that traffic in the Arf6pathway and in large vacuoles in cells expressing the Arf6dominant-active mutant. These results show that PrPc isinternalized in a clathrin-independent pathway that is associatedwith Arf6.

Supplementary material available online athttp://jcs.biologists.org/cgi/content/full/122/22/4062/DC1

Key words: Prion, Clathrin, Raft, Arf6

Summary

Clathrin-independent internalization of normal cellularprion protein in neuroblastoma cells is associatedwith the Arf6 pathwayYoung-Shin Kang, Xiaohong Zhao, Jenna Lovaas, Evan Eisenberg and Lois E. Greene*Laboratory of Cell Biology, NHLBI, NIH, Bethesda, MD 20892-0301, USA*Author for correspondence (greenel@helix.nih.gov)

Accepted 24 August 2009Journal of Cell Science 122, 4062-4069 Published by The Company of Biologists 2009doi:10.1242/jcs.046292

Jour

nal o

f Cel

l Sci

ence

4063Endocytosis of PrPc in N2a cells

caveolae-like domains; caveolae are a subset of lipid rafts containingthe protein caveolin. Ying et al. found that endogenous PrPc waslocalized to caveolae in mouse N2a cells (Ying et al., 1992), althoughother laboratories have not even detected the presence of caveolinin N2a cells (Marella et al., 2002; Shyng et al., 1994). Immunogoldlabeling also localized PrPc to caveolae in a CHO cell line that stablyexpresses hamster PrPc (Peters et al., 2003), and endogenous PrPc

has also been detected in lipid rafts, as well as in clathrin-coatedpits, in N2a cells (Sunyach et al., 2003). In addition, PrPc has beenfound in detergent-insoluble extracts from N2a cells, which isindicative of its presence in raft and/or caveolae-like domains(Sunyach et al., 2003; Vey et al., 1996). These raft and/or caveolaestructures seem to be important for PrPc trafficking, because theirdisruption by the use of sterol-binding agents, such as filipin andnystatin, inhibits Cu2+-stimulated PrPc internalization in humanmicroglia cells and N2a cells overexpressing mouse PrPc (Marellaet al., 2002; Peters et al., 2003). Although these results indicate thepresence of PrPc in raft and/or caveolae structures, it has beensuggested that PrPc normally occurs in rafts but then exits the raftsto be internalized via clathrin-coated pits (Morris et al., 2006;Sunyach et al., 2003).

In summary, although there is no consensus, it is generallyconsidered that, at least in neuronal cells, PrPc occurs in rafts, butis then internalized via clathrin-mediated endocytosis. To investigatethe importance of this latter pathway for the internalization of PrPc,we depleted N2a cells of clathrin using RNA interference andmeasured the internalization of the endogenous PrPc. Surprisingly,after clathrin depletion, confirmed by marked inhibition oftransferrin uptake, we found no significant decrease ininternalization of PrPc either in the presence or absence of Cu2+.By contrast, when rafts were disrupted by filipin or nystatin, PrPc

internalization was blocked both in the presence and absence ofCu2+, whereas clathrin-mediated uptake of transferrin wasunaffected. The internalized PrPc colocalized with cargo that wasassociated with the clathrin-independent Arf6-associated pathwayand was present in large vacuoles in cells expressing the Arf6dominant-active mutant. These results show that PrPc is internalizedin a clathrin-independent pathway that is associated with Arf6.

ResultsEffect of clathrin depletion on the internalization of PrPc

It was previously shown that the rates of PrPc and transferrininternalization were the same in primary neurons in the absence ofCu2+ (Sunyach et al., 2003). This suggested that PrPc wasinternalized via clathrin-mediated endocytosis, the pathway fortransferrin internalization. Because N2a cells have been used as amodel to study PrPc trafficking and scrapie propagation, weexamined whether transferrin and PrPc are internalized at the samerate in N2a cells. After loading the plasma membrane with eitherAlexa-Fluor-488–transferrin or anti-PrP Fab at 4°C for 30 minutes,cells were incubated for varying times at 37°C both in the presenceand absence of Cu2+. The fluorescence intensity of PrPc andtransferrin in the cell was measured to determine the rate ofinternalization. The data, plotted in Fig. 1 as fraction internalizedversus time, shows that Cu2+ did not appreciably affect the kineticsof internalization. The major effect of Cu2+ was that it changed thedistribution of PrPc in the cell. In the presence of Cu2+, the internaland external pools of PrPc became equal at steady state, whichindicates that the in and out rates are equal. Because steady statewas reached with a half-life of 5 minutes in the presence of Cu2+,half-lives of 10 minutes were calculated for both internalization

and externalization of PrPc. In the absence of Cu2+, the internalpool was only one-quarter the size of the external pool at steadystate (compared with 50% in the presence of Cu2+), which isconsistent with the stimulatory role of Cu2+ on PrPc internalization(Pauly and Harris, 1998). Because steady state was reached with ahalf-life of about 10 minutes in the absence of Cu2+, half-lives of48 and 12 minutes were calculated for the internalization andexternalization of PrPc, respectively. As for transferrin, it wasinternalized at about the same rate in both the presence and absenceof Cu2+ (Fig. 1B). Because more than 80% of the transferrin receptorwas found in the internal pool, the half-life to reach steady state, 5minutes, is essentially the half-life of internalization. These resultsshow that, in N2a cells, transferrin is internalized about tenfold fasterthan PrPc in the absence of Cu2+, unlike in primary sensory neuronsin which these rates were similar (Sunyach et al., 2003). In thepresence of Cu2+, the in rate of transferrin internalization is onlytwice that of PrPc in N2a cells, but similar measurements have notbeen made in primary neurons.

Fig. 1. Kinetics of internalization of PrPc and transferrin in N2a cells. Cellswere preloaded with anti-PrP Fab or transferrin at 4°C, followed by incubatingthem at 37°C for different time periods either in the absence (white symbols)or presence (black symbols) of Cu2+. (A)The fraction of PrPc internalizedplotted as a function of time. (B)The fraction of transferrin internalized plottedas a function of time. The transferrin data were corrected for the backgroundlevels at the zero time point, which was less than 5% of the fluorescenceintensity of the 30-minute time point. There was no detectable PrPc

internalized at the zero time point, and data were corrected for fluorescencefrom the secondary antibody alone, which was less than 20% of thefluorescence intensity of the 30-minute time point. These graphs werenormalized by taking the ratio of the amount internalized at a given time to theamount measured at the 30-minute time point. The amount of internalized PrPc

or transferrin was calculated from fluorescence-intensity measurements, aftertaking confocal z-stack images of the cells. Fluorescence intensity was alsoused to calculate the size of the internal and external pool of PrPc andtransferrin receptor at steady state. n100 cells for each time point. The graphswere normalized to 1.0, which is the distribution of PrPc or transferrin receptorat steady state. At steady state, the internal pool of PrPc is 25% of the totalcellular PrPc pool in the absence of Cu2+ and 50% of the total cellular PrPc

pool in the presence of Cu2+. For transferrin, the internal pool of transferrinreceptor is 80% of the total receptor in the cell.

Jour

nal o

f Cel

l Sci

ence

4064

These results suggest that, at least in N2a cells in the absence ofCu2+, transferrin and PrPc are not internalized via the same pathwaybecause their internalization rates differ by about one order ofmagnitude. Therefore, to directly test whether, like transferrin, PrPc

is internalized via clathrin-mediated endocytosis, N2a cells weredepleted of clathrin using RNA interference. Knock down of clathrinhas been used to identify cargos that are internalized via clathrin-mediated endocytosis while, at the same time, helping to elucidatealternative internalization pathways. N2a cells were depleted ofclathrin using two different siRNA oligonucleotides, resulting in a95% reduction in the clathrin levels of the siRNA-treated cellscompared with control cells (supplementary material Fig. S1).

The effect of clathrin depletion on PrPc internalization was firstmeasured in the absence of Cu2+. In these experiments, PrPc andtransferrin internalization were measured concomitantly becausetransferrin internalization is a way to monitor clathrin-mediatedendocytosis. In all of our experiments, internalization of PrPc wasmeasured using anti-PrP Fab antibodies, whereas fluorescentlylabeled transferrin was used to measure transferrin uptake.Fluorescent images of PrPc and transferrin internalization, alongwith clathrin immunostaining, are shown for control cells in Fig.2A-C. The clathrin-depleted and control cells were imaged usingidentical confocal settings to enable direct comparison of fluorescentintensity. Consistent with the western blot analysis, the clathrin-depleted cells (Fig. 2D-F) showed a marked reduction in clathrinstaining at the trans-Golgi network and a concomitant decrease intransferrin uptake. These results showed that clathrin-siRNAtreatment effectively blocks clathrin-mediated endocytosis.Surprisingly, although transferrin uptake was blocked, the clathrin-depleted cells did not show a significant reduction in PrPc

internalization.In order to quantify the pool of internalized PrPc, the staining

protocol was modified to enable us to clearly resolve internal fromsurface-bound PrPc. The surface bound anti-PrP Fab was stainedwith a Cy5-conjugated secondary antibody prior to cellpermeabilization, whereas the internal pool of PrPc was labeled witha rhodamine-conjugated secondary antibody after permeabilization.Fig. 3A shows immunostaining of PrPc in control and clathrin-depleted cells using this method, along with transferrin uptake. Thesurface PrPc outlining the cell is now clearly resolved from theinternal pool. These images confirm that clathrin depletion caused

Journal of Cell Science 122 (22)

marked inhibition of transferrin uptake, while not significantlyaffecting PrPc internalization.

To analyze the effect of clathrin depletion on PrPc and transferrininternalization, more than 150 control and clathrin-depleted cellswere randomly imaged. The average intensities of PrPc andtransferrin in the recycling endosome were then measured in eachcell. In Fig. 3B, the values measured in the clathrin-depleted cellsare normalized to the control values. This graph shows that therewas no significant difference in the average intensity of PrPc in therecycling endosome between control and clathrin-depleted cells,whereas transferrin uptake was reduced by 85% in clathrin-depletedcells compared with controls. Cells were also depleted usinganother clathrin-siRNA oligonucleotide (oligonucleotide #2) toensure that the particular siRNA oligonucleotide used was nothaving an effect on PrPc internalization. With oligonucleotide #2,not only was the data analyzed using the average intensity of the

Fig. 2. Internalization of transferrin and PrPc in control and clathrin-depletedcells measured in the absence of Cu2+. Control (A-C) and clathrin-depleted(D-F) cells were imaged for the following proteins: transferrin (Tfn; A,D),PrPc (B,E), and clathrin (C,F). Identical confocal settings were used in imagingcontrol and clathrin-depleted cells. Internalization of PrPc was measured usingD13 anti-PrP Fab, whereas transferrin was measured using Alexa-Fluor-488–transferrin. Cells were fixed, permeabilized, and then immunostained forclathrin. Secondary anti-Fab was added at the same time to detect PrPc.

Fig. 3. Effect of clathrin depletion on the constitutive internalization of PrPc.(A)Control (a-c) and clathrin-depleted (d-f) cells were simultaneously imagedfor surface PrPc (a,d), internalized PrPc (b,e) and transferrin (c,f).Internalization was performed in the absence of Cu2+. (B)Graph ofinternalized PrPc (lanes 1, 2, 5, 6) and transferrin (Tfn; lanes 3, 4, 7, 8)measured in control (lanes 1, 3, 5, 7) and clathrin-siRNA-treated (lanes 2, 4, 6,8) cells. Oligonucleotide #1 (lanes 2, 4) and oligonucleotide #2 (lanes 6, 8)were used to deplete cells of clathrin. For lanes 1-4, internalization oftransferrin and PrPc was measured by determining the average intensity ofthese proteins in the recycling endosome using a large pinhole when imagingthe cells. n150 cells per data set. For lanes 5-8, internalization of transferrinand PrPc represents the internal pools of internalized transferrin and PrPc

calculated from z-stacks of the cells using MetaMorph software to analyze theimages. n12-17 cells per data set. The measured PrPc and transferrin in theclathrin-depleted cells were normalized to control cells.

Jour

nal o

f Cel

l Sci

ence

4065Endocytosis of PrPc in N2a cells

recycling endosome, the internalization of PrPc and transferrin wasalso analyzed using MetaMorph software to determine the totalinternal pools of PrPc and transferrin from confocal z-stack cellimages. The results from the three-dimensional analysis were notsignificantly different from the two-dimensional analysis, i.e.clathrin depletion did not significantly affect the extent of PrPc

internalization, but markedly inhibited transferrin uptake (Fig. 3B,lanes 5-8). Therefore, constitutive internalization of PrPc apparentlyoccurs via a clathrin-independent pathway.

Similar experiments were carried out in the presence of Cu2+,which has been shown to stimulate the internalization of PrPc (Paulyand Harris, 1998). Fig. 4A shows the immunostaining of surfaceand internal pools of PrPc, along with transferrin uptake, in cellsthat were incubated in the presence of Cu2+. Comparison of PrPc

localization in the presence and absence of Cu2+ showed that, inaddition to a larger internal pool of PrPc in the presence of Cu2+,the PrPc on the plasma membrane also had a patchier appearancecompared with in the absence of Cu2+. Just as we observed in theabsence of Cu2+, clathrin depletion of the N2a cells did notsignificantly affect internalization of PrPc (Fig. 4Ac-f), although,as expected, the clathrin-depleted cells showed a marked reductionin transferrin uptake. Again, we used two different siRNAoligonucleotides to deplete clathrin and two slightly differentmethods to measure the internalization of proteins. This quantitativeanalysis, shown in Fig. 4B, confirmed that clathrin depletion didnot significantly affect the Cu2+-stimulated PrPc endocytic pathway,although it markedly inhibited transferrin uptake. Therefore, PrPc

is internalized via a clathrin-independent pathway both in thepresence and absence of Cu2+.

All of the above PrPc-internalization studies examining the effectof clathrin depletion were performed using the D13 anti-PrP Fab.To validate these results further, PrPc internalization was remeasuredusing a different anti-PrP Fab, one made from the AH6 anti-PrPmAb. This latter antibody recognizes residues 159-174, whereasthe D13 Fab recognizes residues 95-105 (Novitskaya et al., 2006).Data obtained with the AH6 Fab confirms that neither constitutivenor Cu2+-stimulated internalization of PrPc was significantly affectedby blocking clathrin-mediated endocytosis (Table 1).

Characterization of the alternative pathway for PrPc

internalizationTo better characterize the clathrin-independent pathway forinternalizing PrPc, we examined the effect of disrupting lipid raftson PrPc internalization. Lipid rafts were disrupted by treating cellswith either filipin or nystatin, drugs that disrupt rafts by sequesteringcholesterol while only having a limited effect on clathrin-mediatedendocytosis (Orlandi and Fishman, 1998; Ricci et al., 2000). Whenthe cells were incubated with nystatin, PrPc internalization wasblocked both in the presence and absence of Cu2+ (Fig. 5A). Bycontrast, transferrin uptake was not significantly affected by nystatintreatment. Quantitative analysis of PrPc and transferrininternalization in nystatin-treated cells showed a marked inhibitionof PrPc internalization and only a slight decrease in transferrinendocytosis (Fig. 5B). Similar results were obtained by treating thecells with filipin (Fig. 5B). In agreement with these data, Marellaet al. found that, in the presence of Cu2+, nystatin and filipin blockedinternalization of PrPc, but did not block transferrin uptake (Marellaet al., 2002). Therefore, disruption of raft organization inhibitedPrPc internalization both in the presence and absence of Cu2+. Inaddition to being dependent on rafts, PrPc internalization was foundto be dynamin dependent both in the presence and absence of Cu2+.When cells were transfected with GFP constructs of either wild-

Fig. 4. Effect of clathrin depletion on Cu2+-stimulated internalization of PrPc.(A)Control (a-c) and clathrin-depleted (d-f) cells were simultaneously imagedfor surface PrPc (a,d), internalized PrPc (b,e), and transferrin (Tfn; c,f).Internalization was performed in the presence of 400M CuSO4. (B)Graph ofinternalized PrPc (lanes 1, 2, 5 6) and transferrin (lanes 3, 4, 7 and 8) measuredin control (lanes 1, 3, 5, 7) and clathrin-siRNA-treated (lanes 2, 4, 6, 8) cells.Oligonucleotide #1 (lanes 2, 4) and oligonucleotide #2 (lanes 6, 8) were usedto deplete cells of clathrin. Internalized transferrin and PrPc were measuredusing the average intensity of these proteins in the recycling endosome forlanes 1-4 (n150 cells per data set) and the total intensity for these proteins inthe recycling endosome for lanes 5-8 (n12-17 cells per data set). Themeasured PrPc and transferrin in the clathrin-depleted cells were normalized tothe control cells.

Table 1. Internalization of PrPc measured using anti-PrP FabAH6 mAb in control and clathrin-depleted cells*

– CuSO4 + CuSO4

PrPc Tfn PrPc Tfn

Control 1.00±0.31 1.00±0.30 1.00±019 1.00±0.30Clathrin siRNA 1.00±0.30 0.18±0.03 1.02±0.21 0.19±0.04

*Internalization of PrPc and transferrin (Tfn) was measured using theaverage intensity of these proteins in the recycling endosome (n10-13 cellsper data set). The measured PrPc and transferrin in the clathrin-depleted cellswere normalized to the control cells.

Jour

nal o

f Cel

l Sci

ence

4066

type dynamin-1 or the dominant-negative mutant of dynamin-1,dynamin-1 (K44A), the latter inhibited internalization of both PrPc

and transferrin in the presence and absence of Cu2+ (Table 2,supplementary material Fig. S2A). Consistent with these results,incubation of the cells with the dynamin inhibitor dynasore alsoinhibited PrPc internalization both in the presence and absence ofCu2+ (Table 2, supplemental material Fig. S2B). Therefore, whetheror not Cu2+ is present, the internalization of PrPc in N2a cells israft- and dynamin-dependent, but clathrin-independent.

A raft-dependent dynamin-dependent pathway would typicallysuggest that PrPc is internalized via a caveolae-raft internalizationpathway. However, the Harris laboratory reported that N2a cellscontain trace amounts of caveolin-1 (Shyng et al., 1994) and weconfirmed their finding using western blot analysis to show thatthe level of caveolin-1 in N2a cells was barely detectable comparedwith the level in HeLa cells (supplemental material Fig. S3A). Toconfirm that the internalization of PrPc does not occur via caveolae,cells were transiently transfected with GFP-labeled caveolin-1 witha mutation in tyrosine 14 (Y14F). Because phosphorylation of

Journal of Cell Science 122 (22)

tyrosine 14 in caveolin is essential for caveolae internalization (delPozo et al., 2005; Orlichenko et al., 2006), mutation of this tyrosineresidue inhibits internalization of caveolin-associated rafts. In fact,we found that expression of this mutant caveolin-1 had no effecton PrPc internalization either in the presence or absence of Cu2+

(supplemental material Fig. S3B). By contrast, PrPc internalizationwas reduced by about 50% when actin was depolymerized bycytochalasin D.

Because nystatin and filipin have been shown to blockinternalization of cargo that traffic via the Arf6-regulated pathway(Naslavsky et al., 2003), we examined whether PrPc colocalizes withother cargo that traffic via this pathway. After internalizing PrPc, thecells were immunostained to detect CD98, a glycoprotein that trafficsvia the Arf6-associated pathway (Eyster et al., 2009). Both in thepresence and absence of Cu2+, internalized PrPc was found tocolocalize with CD98 in vacuolar structures that are frequently foundin N2a cells (Fig. 6A). As expected, no internalized transferrin wasassociated with PrPc in these decorated vacuolar structures(supplemental material Fig. S4). We next determined whether cargothat traffics via the Arf6 pathway and is internalized at the same timeas PrPc also colocalizes with PrPc. For this experiment, we followedthe internalization of MHC-I, another protein whose trafficking viathe Arf6 pathway has been well characterized (Naslavsky et al., 2003).Cells were preloaded with anti-PrP and anti-MHC-I antibodies,followed by incubation at 37°C for 30 minutes. Following stainingof these proteins with secondary antibodies, MHC-I and PrPc werefound to be colocalized on the same vacuolar structures both in thepresence and absence of Cu2+ (Fig. 6B).

To further validate that PrPc trafficking is associated with theArf6 pathway, N2a cells were transfected with the dominant-positiveArf6 mutant Arf6Q67L. This mutant has been shown to drive theformation of vacuoles that trap cargo that enters cells via a clathrin-independent pathway prior to the cargo merging with the clathrinpathway (Donaldson et al., 2009). If trafficking of PrPc wasregulated by Arf6, then PrPc would be found on the large vacuolesthat develop in cells expressing Arf6Q67L (Brown et al., 2001).As shown in Fig. 7A, both in the presence and absence of Cu2+,internalized PrPc indeed decorated large vacuoles that co-stainedfor CD98. Moreover, when antibodies against PrPc and MHC-I wereused to simultaneously examine the uptake of these proteins, PrPc

and MHC-I were colocalized on large vacuoles both in the presenceand absence of Cu2+ (Fig. 7B). PrPc was found localized to theselarge vacuoles even in cells incubated with the dynamin inhibitordynasore.

Fig. 5. Effect of the disruption of lipid rafts on the internalization of PrPc in thepresence and absence of Cu2+. (A)Nystatin-treated cells were imaged forsurface PrPc (a,d), internalized PrPc (b,e) and transferrin (Tfn; c,f) in theabsence (a-c) and presence (d-f) of Cu2+. (B)Plot of PrPc and transferrininternalization in nystatin- or filipin-pretreated cells incubated in the absence(lanes 1-6) and presence (lanes 7-12) of Cu2+. The amount of internalized PrPc

and transferrin was quantified in control (lanes 1, 4, 7, 10), nystatin-treated (2,5, 8, 11) and filipin-treated (lanes 3, 6, 9, 12) cells. Internalization oftransferrin and PrPc was calculated based on the total intensity of theseproteins in the recycling endosome (n12-18 cells per data set). The measuredPrPc and transferrin in the nystatin-treated and filipin-treated cells werenormalized to the control cells.

Table 2. Internalization of PrPc is dynamin-dependent*

– CuSO4 + CuSO4

PrPc Tfn PrPc Tfn

Control 1.00±0.17 1.00±0.14 1.00±0.04 1.00±0.07WT-Dyn1 1.00±0.17 1.00±0.07 1.00±0.14 1.00±0.13K44A Dyn1 0.17±0.08 0.07±0.03 0.19±0.09 0.07±0.03Dynasore† 0.25±0.05 0.10±0.05 0.20±0.04 0.08±0.05

*Internalization of PrPc and transferrin (Tfn) was measured using theaverage intensity of these proteins in the recycling endosome (n10-13 cellsper data set). The measured PrPc and transferrin in the cells expressing WT-Dyn1 or K44A Dyn1 and in Dynasore-treated cells were normalized to thecontrol cells.

†Cells were incubated with 10 g/ml dynasore for 30 minutes prior tomeasuring internalization.

WT-Dyn1, wild-type dynamin-1.

Jour

nal o

f Cel

l Sci

ence

4067Endocytosis of PrPc in N2a cells

To determine whether other cargos are found on these largevacuolar structures, the internalization of cholera toxin B andtransferrin was examined in cells expressing Arf6Q67L. Both inthe presence and absence of Cu2+, cholera toxin B was found to becolocalized with CD98 on these large vacuolar structures (Fig. 8).These results indicate that the trafficking of cholera toxin B isassociated with the Arf6 pathway in N2a cells; this result has alsobeen reported for the trafficking of cholera toxin B in HeLa cells(Naslavsky et al., 2004). As expected, there was no significantcolocalization of internalized transferrin with CD98-immunostainedvacuoles (Fig. 8). Therefore, the Arf6Q67L vacuoles trap PrPc andcholera toxin B, but not transferrin. These results show that PrPc

is internalized via a clathrin-independent pathway that is associatedwith Arf6 in N2a cells.

DiscussionTo determine whether clathrin-mediated endocytosis is essential forinternalizing PrPc, RNA interference was used to deplete clathrinin N2a cells. This technique has been extensively used to identifythe role of clathrin in the internalization of various cargos. Severalstudies have indicated that PrPc is internalized via clathrin-mediatedendocytosis in neuronal cells (Shyng et al., 1994; Sunyach et al.,2003; Taylor et al., 2005). Therefore, we expected that clathrindepletion would block PrPc internalization. Surprisingly, both inthe presence and absence of Cu2+, the same amount of PrPc was

internalized in cells treated with clathrin siRNA as in control cells.Although blocking the clathrin-dependent pathway did notsignificantly affect PrPc internalization, disrupting lipid rafts withnystatin or filipin markedly inhibited the internalization of PrPc bothin the presence and absence of Cu2+ without significantly affectingtransferrin uptake. These results suggest that, at least in N2a cells,both in the presence and absence of Cu2+, endogenous PrPc isinternalized via a clathrin-independent pathway.

Our results are in agreement with the study from the Chabrylaboratory showing inhibition of PrPc internalization by filipin,which disrupts rafts, but not by chloroperazine, which blocksclathrin-mediated endocytosis (Marella et al., 2002). Conversely,our findings disagree with a number of other studies that suggestthat PrPc is internalized via clathrin-mediated endocytosis (Shynget al., 1994; Taylor et al., 2005; Sunyach et al., 2003). Thisdisagreement might partly be due to differences in cell types, aswell as to differences in trafficking of endogenous PrPc and stablyexpressed PrPc. In addition, instead of using pharmacological agentswith ill-defined targets to block clathrin, our study blocked clathrininternalization using siRNA, which specifically inhibits clathrin-mediated endocytosis. Finally, support for the view that PrPc isinternalized via clathrin-mediated endocytosis has come fromelectron-microscopic images showing immunogold-labeled PrPc inclathrin-coated pits (Shyng et al., 1994; Sunyach et al., 2003), butthese images do not provide a measure of the extent to which PrPc

Fig. 6. Internalized PrPc localized with cargo thattraffic via the Arf6-associated pathway.(A)Following internalization of PrPc for 30minutes, cells were fixed and then co-stained forCD98. (B)Following internalization of PrPc andMHC-I for 30 minutes, cells were immunostainedwith secondary antibodies. Internalization ofMHC-I was conducted by preloading the cells withanti-MHC-I antibody at 4°C, followed byincubation at 37°C. Scale bars: 10m.

Fig. 7. PrPc colocalizes with CD98 and MHC-I invacuoles in cells expressing the dominant-positiveArf6 mutant Arf6Q67L. (A)Following transfection ofcells with Arf6Q67L, PrPc was internalized for 30minutes, then cells were fixed and immunostained forCD98. (B)Following transfection of cells withArf6Q67L, PrPc and MHC-I were internalized for 30minutes, fixed and stained with secondary antibodies.Scale bars: 10m.

Jour

nal o

f Cel

l Sci

ence

4068

traffics via this pathway. In fact, electron microscopy of theimmunolabeled PrPc found that greater than 80% of the labeledPrPc was in detergent-resistant membranes (Morris et al., 2006).

PrPc belongs to the family of GPI-anchored proteins, which aretypically internalized via clathrin-independent pathways includinga caveolin-pathway of uptake, flotillin-1 pathway, a Cdc42-regulatedpathway and an Arf6-regulated pathway (Donaldson et al., 2009;Glebov et al., 2006; Mayor and Riezman, 2004). Our results usingthe dominant-active Arf6 mutant, in combination with our resultsusing clathrin siRNA, show that PrPc traffics via a clathrin-independent Arf6-associated pathway. In addition to the presenceof PrPc in Arf6Q67L vacuoles, the internalized PrPc colocalized withCD98 and MHC-I, proteins that traffic via the Arf6-regulatedpathway (Eyster et al., 2009; Naslavsky et al., 2003). Typically, cargothat traffics via the Arf6 pathway is dynamin independent (Donaldsonet al., 2009), but perhaps this is different in neuronal cells, whichexpress both neuronal-specific dynamin-1 as well as dynamin-2.

It is important to recognize that many cargos are internalized viamultiple pathways. For example, similar to PrPc in our study, the-amyloid peptide is internalized in a clathrin-independent,dynamin-dependent and raft-dependent pathway in the absence ofapoliproprotein E (Saavedra et al., 2007). By contrast, in thepresence of apoliproprotein E, -amyloid peptide is internalized viaclathrin-mediated endocytosis (Kang et al., 2000). Furthermore,even though there is disagreement on whether PrPc is internalizedvia a clathrin-dependent or -independent pathway, it is importantto note that both of these pathways converge either directly orindirectly on the early endosome (Donaldson et al., 2009; Kirkhamet al., 2005). It has previously been shown that PrPc first traffics tothe early endosome (Magalhaes et al., 2002; Pimpinelli et al., 2005)and then to the late endosome (Peters et al., 2003; Pimpinelli et al.,2005). Ultimately, it is then either degraded in the lysosome orrecycled back to the plasma membrane (Borchelt et al., 1992).

Although the simplest way of interpreting our studies that usedclathrin siRNA is that the majority of PrPc does not enter the cellsvia clathrin-mediated endocytosis, it is possible that, in N2a cells,PrPc is internalized via multiple pathways. Perhaps when clathrinis depleted, a clathrin-independent raft pathway is upregulated.

Journal of Cell Science 122 (22)

Compensatory mechanisms of trafficking were previously observedwhen a dominant-negative dynamin-1 mutant, dynamin-1 K44A,was expressed in cells (Damke et al., 1994). Therefore, it is possiblethat a clathrin-independent pathway compensates for the clathrin-mediated endocytosis when clathrin is depleted. As for the inhibitionof PrPc internalization that occurs in the presence of clathrin whenthe raft pathway is disrupted, this might occur if PrPc must be presentin intact rafts before it can shift over to clathrin-coated pits to beinternalized via clathrin-mediated endocytosis (Taylor and Hooper,2006). One way to test this rather complicated model would be toisolate the carrier protein that PrPc binds to during internalization.A recent report from the Hooper laboratory suggested that the LRP1receptor was the carrier for PrPc (Taylor and Hooper, 2007), but amore recent study from the Morris laboratory suggested that theprimary role of this receptor was to transport PrPc to the plasmamembrane rather than act as a carrier for internalization of PrPc

(Parkyn et al., 2008). Isolation of the PrPc carrier protein mightfinally provide a definitive answer as to how PrPc is internalizedin N2a cells and primary neurons.

Materials and MethodsCell culture and transfectionThe N2a mouse neuroblastoma cell line was purchased from ATCC (American TypeCulture Collection). N2a cells were grown at 37°C in a humidified incubator under95% air, 5% CO2 in a MEM (Mediatech, Herndon, VA) supplemented with 10%FBS, 100 IU/ml penicillin, 200 g/ml streptomycin and nonessential amino acids(Invitrogen, Carlsbad, CA). Cells, cultured for 24 hours in complete medium withoutantibiotics, were transfected with the following constructs using Lipofectamine 2000(Invitrogen, Carlsbad, CA): GFP-labeled wild-type dynamin-1 and K44A dynamin-1 (gift from Pietro De Camilli, Yale University, New Haven, CT), Arf6Q67L (giftfrom Julie Donaldson, NHLBI, NIH, Bethesda, MD), and GFP-labeled caveolin-1and GFP-labeled mutant Y14F caveolin-1 (gift from Mark A. McNiven, Mayo Clinic,Rochester, MN).

Internalization of PrPc

N2a cells, grown on eight-well-chamber glass slides (Nalge Nunc International,Rochester, NY), were washed twice with phosphate buffer saline and then incubatedat 4°C for 30 minutes with 2.5 g/ml anti-PrP Fab antibody. Transferrin (10 g/ml),either Alexa-Fluor-488-conjugated transferrin (Invitrogen, Carlsbad, CA) or Cy5-conjugated transferrin (Jackson ImmunoResearch, West Grove, PA), was routinelyadded to the cells, followed by incubation of the cells for 30 minutes at 37°C priorto fixation. To follow cholera-toxin-B internalization, rhodamine-labeled cholera toxinB (Sigma) was added at 1 g/ml. In the experiments in the presence of Cu2+, 400 MCuSO4 was added to the cells when they were incubated at 37°C. In the lipid-raft-disruption experiments, cells were first treated nystatin (50 g/ml) and filipin (5 g/ml)(Sigma-Aldrich, St Louis, MO) in complete medium at 37°C for 30 minutes. Cellscontinued to be incubated with the drugs during the preincubation at 4°C andincubation at 37°C. To distinguish between internal and surface-bound PrPc, fixedcells were first immunostained with Cy5-conjugated secondary anti-Fab antibodyand then, following permeabilization, were immunostained with a secondary anti-Fab antibody conjugated to a different fluorophore to detect internalized PrPc. Todepolymerize actin, 1 M cytochlasin D (Sigma) was added to N2a cells for 30 minutesat 37°C. To inhibit dynamin, 10 M dynasore (Sigma) was added to cells at 37°Cfor 30 minutes.

AntibodiesThe anti-PrP Fab antibody used was either D13 Fab (InPro, San Francisco, CA) orAH6 Fab, made from the AH6 antibody (TSE Resource Center, Berkshire, UK) usingthe Fab preparation kit (Pierce, Rockford, IL). Cells were immunostained for clathrinwith X22 antibody (Affinity BioReagents, Golden, CO), GFP antibody (Abcam) andAlexa-Fluor-488-conjugated CD98 antibody (AbD Serotec, Raleigh, NC).Immunoblots were probed with anti-clathrin-heavy-chain mAb clone 23 (BDBiosciences, San Jose, CA), -actin (Abcam, Cambridge MA), caveolin-1 (SantaCruz Biotechnology, Santa Cruz, CA), glycerol 3-phosphate dehydrogenase (NovusBiologicals, Littleton, CO) and major histocompatibility protein class I (MHC-I) (fromNatalie Porat-Shoram, NHLBI, NIH, Bethesda, MD). Fluorescent secondaryantibodies were from Jackson ImmunoResearch and Invitrogen.

Immunofluorescence microscopy and analysisImages were obtained using the LSM510 confocal microscope (Zeiss, Thornwood,NY) using Plan-Apochromat 63�/1.4 objective. In comparing control and eithersiRNA- or drug-treated cells, identical confocal settings were used to enable direct

Fig. 8. Internalized cholera toxin B, but not internalized transferrin,colocalizes with CD98 vacuoles in cells expressing the dominant-positive Arf6mutant Arf6Q67L. Following transfection of cells with Arf6Q67L, choleratoxin (A,A�) and transferrin (B,B�) was internalized for 30 minutes, then cellswere fixed and immunostained for CD98 (C,C�) in the absence (A-C) andpresence (A�-C�) of Cu2+. Scale bars: 10m.

Jour

nal o

f Cel

l Sci

ence

4069Endocytosis of PrPc in N2a cells

comparison between images. Analysis of the internalized PrPc and transferrin wasperformed using MetaMorph software (Molecular Devices, Sunnyvale, CA).

Clathrin siRNARNA interference of clathrin heavy chain was performed using siRNAs (Dharmacon,Chicago, IL) to the following sequence: 5�-UCAGAAGAAUUGCUGC-UUAUU-3� (oligonucleotide #1) and 5�-UAAUCCAAUUCGAAGACCAAUUU-3�(oligonucleotide #2). Lipofectamine RNAiMAX (Invitrogen) was used to transfectthe siRNA oligonucleotides using both the forward and reverse method to transfectcells as described in the manufacturer’s transfection protocol. To determine cellularlevels of clathrin, cell lysates were run on 4-12% gels (Invitrogen), followed bytransferring the proteins to nitrocellulose for immunoblotting. Protein bands weredetected using the Odyssey infrared detection system (Li-Cor Bioscience, Lincoln,NE) or chemiluminescent substrate (Pierce, cat. no. 34080) followed by densitometerimaging (ChemiImager, Alpha Innotech).

We thank Julie Donaldson and members of the Donaldson laboratoryfor helpful discussions.

ReferencesAguzzi, A., Raeber, A., Blattler, T., Flechsig, E., Klein, M., Weissmann, C. and

Brandner, S. (1997). Neurotoxicity and neuroinvasiveness of prions. J. Neurovirol. 3Suppl. 1, S23-S24.

Borchelt, D. R., Taraboulos, A. and Prusiner, S. B. (1992). Evidence for synthesis ofscrapie prion proteins in the endocytic pathway. J Biol. Chem. 267, 16188-16199.

Brown, F. D., Rozelle, A. L., Yin, H. L., Balla, T. and Donaldson, J. G. (2001).Phosphatidylinositol 4,5-bisphosphate and Arf6-regulated membrane traffic. J. Cell Biol.154, 1007-1018.

Collinge, J., Owen, F., Poulter, M., Leach, M., Crow, T. J., Rossor, M. N., Hardy, J.,Mullan, M. J., Janota, I. and Lantos, P. L. (1990). Prion dementia without characteristicpathology. Lancet 336, 7-9.

Damke, H., Baba, T., Warnock, D. E. and Schmid, S. L. (1994). Induction of mutantdynamin specifically blocks endocytic coated vesicle formation. J. Cell Biol. 127, 915-934.

del Pozo, M. A., Balasubramanian, N., Alderson, N. B., Kiosses, W. B., Grande-Garcia,A., Anderson, R. G. W. and Schwartz, M. A. (2005). Phospho-caveolin-1 mediatesintegrin-regulated membrane domain internalization. Nat. Cell. Biol. 7, 901.

Donaldson, J. G., Porat-Shliom, N. and Cohen, L. A. (2009). Clathrin-independentendocytosis: A unique platform for cell signaling and PM remodeling. Cell. Sig. 21, 1-6.

Eyster, C. A., Higginson, J. D., Huebner, R., Porat-Shliom, N., Weigert, R., Wu, W.W., Shen, R.-F. and Donaldson, J. G. (2009). Discovery of new cargo proteins thatenter cells through clathrin-independent endocytosis. Traffic 10, 590-599.

Glebov, O. O., Bright, N. A. and Nichols, B. J. (2006). Flotillin-1 defines a clathrin-independent endocytic pathway in mammalian cells. Nat. Cell. Biol. 8, 46.

Kang, D. E., Pietrzik, C. U., Baum, L., Chevallier, N., Merriam, D. E., Kounnas, M.Z., Wagner, S. L., Troncoso, J. C., Kawas, C. H., Katzman, R. et al. (2000). Modulationof amyloid beta-protein clearance and Alzheimer’s disease susceptibility by the LDLreceptor-related protein pathway. J Clin. Invest. 106, 1159-1166.

Kirkham, M., Fujita, A., Chadda, R., Nixon, S. J., Kurzchalia, T. V., Sharma, D. K.,Pagano, R. E., Hancock, J. F., Mayor, S. and Parton, R. G. (2005). Ultrastructuralidentification of uncoated caveolin-independent early endocytic vehicles. J. Cell Biol.168, 465-476.

Magalhaes, A. C., Silva, J. A., Lee, K. S., Martins, V. R., Prado, V. F., Ferguson, S. S.G., Gomez, M. V., Brentani, R. R. and Prado, M. A. M. (2002). Endocyticintermediates involved with the intracellular trafficking of a fluorescent cellular prionprotein. J. Biol. Chem. 277, 33311-33318.

Marella, M., Lehmann, S., Grassi, J. and Chabry, J. (2002). Filipin prevents pathologicalprion protein accumulation by reducing endocytosis and inducing cellular PrP release.J. Biol. Chem. 277, 25457-25464.

Mayor, S. and Riezman, H. (2004). Sorting GPI-anchored proteins. Nat. Rev Mol. Cell.Biol. 5, 110.

Medori, R., Tritschler, H. J., LeBlanc, A., Villare, F., Manetto, V., Chen, H. Y., Xue,R., Leal, S., Montagna, P., Cortelli, P. et al. (1992). Fatal familial insomnia, a priondisease with a mutation at codon 178 of the prion protein gene. N. Engl. J. Med. 326,444-449.

Morris, R. J., Parkyn, C. J. and Jen, A. (2006). Traffic of prion protein between differentcompartments on the neuronal surface, and the propagation of prion disease. FEBS letters580, 5565.

Naslavsky, N., Weigert, R. and Donaldson, J. G. (2003). Convergence of non-clathrin-and clathrin-derived endosomes involves Arf6 inactivation and changes inphosphoinositides. Mol. Biol. Cell 14, 417-431.

Naslavsky, N., Weigert, R. and Donaldson, J. G. (2004). Characterization of a nonclathrinendocytic pathway: membrane cargo and lipid requirements. Mol. Biol. Cell 15, 3542-3552.

Novitskaya, V., Makarava, N., Bellon, A., Bocharova, O. V., Bronstein, I. B., Williamson,R. A. and Baskakov, I. V. (2006). Probing the conformation of the prion protein withina single amyloid fibril using a novel immunoconformational assay. J. Biol. Chem. 281,15536-15545.

Orlandi, P. A. and Fishman, P. H. (1998). Filipin-dependent inhibition of cholera toxin:evidence for toxin internalization and activation through caveolae-like domains. J. CellBiol. 141, 905-915.

Orlichenko, L., Huang, B., Krueger, E. and McNiven, M. A. (2006). Epithelial growthfactor-induced phosphorylation of caveolin 1 at tyrosine 14 stimulates caveolae formationin epithelial cells. J. Biol. Chem. 281, 4570-4579.

Parkyn, C. J., Vermeulen, E. G. M., Mootoosamy, R. C., Sunyach, C., Jacobsen, C.,Oxvig, C., Moestrup, S., Liu, Q., Bu, G., Jen, A. et al. (2008). LRP1 controls biosyntheticand endocytic trafficking of neuronal prion protein. J. Cell Sci. 121, 773-783.

Pauly, P. C. and Harris, D. A. (1998). Copper stimulates endocytosis of the prion protein.J. Biol. Chem. 273, 33107-33110.

Peters, P. J., Mironov, A., Jr, Peretz, D., van Donselaar, E., Leclerc, E., Erpel, S.,DeArmond, S. J., Burton, D. R., Williamson, R. A., Vey, M. et al. (2003). Traffickingof prion proteins through a caveolae-mediated endosomal pathway. J. Cell Biol. 162,703-717.

Pimpinelli, F., Lehmann, S. and Maridonneau-Parini, I. (2005). The scrapie prion proteinis present in flotillin-1-positive vesicles in central- but not peripheral-derived neuronalcell lines. Eur. J. Neurosci. 21, 2063-2072.

Prusiner, S. B. (1998). Prions. Proc. Natl. Acad. Sci. USA 95, 13363-13383.Ricci, V., Galmiche, A., Doye, A., Necchi, V., Solcia, E. and Boquet, P. (2000). High

cell sensitivity to Helicobacter pylori VacA toxin depends on a GPI-anchored proteinand is not blocked by inhibition of the clathrin-mediated pathway of endocytosis. Mol.Biol. Cell 11, 3897-3909.

Saavedra, L., Mohamed, A., Ma, V., Kar, S. and de Chaves, E. P. (2007). Internalizationof beta-amyloid peptide by primary neurons in the absence of apolipoprotein E. J. Biol.Chem. 282, 35722-35732.

Shyng, S. L., Heuser, J. E. and Harris, D. A. (1994). A glycolipid-anchored prion proteinis endocytosed via clathrin-coated pits. J. Cell Biol. 125, 1239-1250.

Sunyach, C., Jen, A., Deng, J., Fitzgerald, K. T., Frobert, Y., Grassi, J., McCaffrey,M. W. and Morris, R. (2003). The mechanism of internalization ofglycosylphosphatidylinositol-anchored prion protein. EMBO J. 22, 3591-3601.

Taylor, D. R. and Hooper, N. M. (2006). The prion protein and lipid rafts (Review). Mol.Memb. Biol. 23, 89-99.

Taylor, D. R. and Hooper, N. M. (2007). The low-density lipoprotein receptor-relatedprotein 1 (LRP1) mediates the endocytosis of the cellular prion protein. Biochem. J.402, 17-23.

Taylor, D. R., Watt, N. T., Perera, W. S. S. and Hooper, N. M. (2005). Assigning functionsto distinct regions of the N-terminus of the prion protein that are involved in its copper-stimulated, clathrin-dependent endocytosis. J. Cell Sci. 118, 5141-5153.

Vey, M., Pilkuhn, S., Wille, H., Nixon, R., DeArmond, S. J., Smart, E. J., Anderson,R. G., Taraboulos, A. and Prusiner, S. B. (1996). Subcellular colocalization of thecellular and scrapie prion proteins in caveolae-like membranous domains. Proc. Natl.Acad. Sci. USA 93, 14945-14949.

Weissmann, C. and Flechsig, E. (2003). PrP knock-out and PrP transgenic mice in prionresearch. Br. Med. Bull. 66, 43-60.

Ying, Y. S., Anderson, R. G. and Rothberg, K. G. (1992). Each caveola contains multipleglycosyl-phosphatidylinositol-anchored membrane proteins. Cold Spring Harb. Symp.Quant. Biol. 57, 593-604.

Jour

nal o

f Cel

l Sci

ence