![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Molecular and Cellular Biology, April 2006, p. 2697-2715, Vol. 26, No. 7
0270-7306/06/$08.00+0 doi:10.1128/MCB.26.7.2697-2715.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Institute of Pathology, Case Western Reserve University, 2085 Adelbert Road, Cleveland, Ohio 44106
Received 18 August 2005/ Returned for modification 6 October 2005/ Accepted 5 January 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this report, we demonstrate the possible mechanisms of cytotoxicity by CtmPrP and the complexities underlying its biogenesis. In designing this study, advantage was taken of the fact that CtmPrP maintains an uncleaved N-terminal signal peptide (N-SP) by virtue of its orientation across the endoplasmic reticulum (ER) lipid bilayer (31). Thus, known antibody epitope tags were inserted in the N-SP of PrP such that the inserted amino acids favor CtmPrP synthesis (22). The 3F4 epitope was inserted before and the Flag epitope following the initiating methionine codon of the PrP sequence (PrPN3F4 and PrPNFlag, respectively) (Fig. 1B). The former is a well-characterized anti-PrP epitope, and the latter has been used extensively for tagging proteins. This approach altered the length and relative charge of the N region of PrP N-SP, two parameters known to favor CtmPrP synthesis (21, 22, 28). The conserved H- and C-terminal domains were left unaltered to preserve the targeting function and cleavage site of the N-SP. The biogenesis of tagged PrP was evaluated in transfected human and mouse neuroblastoma (M17 and N2a, respectively) cells and in Chinese hamster ovary (CHO) cells.
|
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cDNA constructs and transfection of cells. Human PrP cDNA cloned into the cytomegalovirus-driven episomal vector Cep4ß was a kind gift of R. B. Petersen (Case Western Reserve University). Methionine residues 109 and 112 were replaced with leucine and valine, respectively, using the QuikChange site-directed mutagenesis kit according to the manufacturer's directions (Stratagene, La Jolla, CA). To insert 3F4 and Flag epitopes in the N-SP, PrP cDNA was amplified using the following 5' primers coding for the NotI restriction site (boldface), 3F4 or Flag epitope (italics), and PrP N-SP sequence (underlined) (the initiating methionine codon follows the NotI site): 5'GAATTCCGCGGCCGCATGGCGAAAACCAACATGAAGCACATGGCGAACCTTGGCTGCTGG (3F4 epitope) and 5'GAATTCCGCGGCCGCATGGACTACAAGGACGACGATGACAAGGCGAACCTTGGCTGCTGG (NFlag epitope). The 3' primer included a BamHI restriction site and nucleotides corresponding to the C terminus of nascent PrP (synthesized by Operon Technologies Inc., Alameda, CA). The amplified product was digested with NotI and BamHI and inserted in corresponding sites in the Cep4ß eukaryotic expression vector. Cells were transfected with the desired construct by using Lipofectamine according to the manufacturer's specifications (Invitrogen, Grand Island, NY). Transfected human and mouse neuroblastoma cells were cultured in Opti-MEM and 5% fetal calf serum, and CHO cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin.
Fluorescence microscopy. Transfected cells were cultured on poly-d-lysine-coated glass coverslips overnight, fixed, and processed for staining or were first permeabilized with Triton X-100 and then processed further as described in a previous report (12). In a typical experiment, the following primary antibodies were used: monoclonal anti-PrP 3F4, 8H4, or 8B4 (1:20) or polyclonal antitubulin and antivimentin (1:40). Fluorescein- or rhodamine-conjugated secondary antibodies were diluted at 1:40. For double-immunostaining experiments, cells were immunoreacted with 8H4 and anti-mouse antibody-fluorescein isothiocyanate (FITC), washed with phosphate-buffered saline (PBS), and incubated with 1 mg/ml of unlabeled anti-mouse immunoglobulin G before subsequent immunoreaction with 3F4 or 8B4 antibodies. Stained cells were incubated with Hoechst 33342 (1 µg/ml) (Molecular Probes) for 5 minutes to detect apoptotic nuclei and mounted in gel-mount (Biomeda Corp., Foster City, CA). All samples were observed with a laser scanning confocal microscope (Bio-Rad MRC 600). A single 5-µm optical section and a composite of several sections were evaluated in each case.
Electron microscopy. For immunoelectron microscopy, PrPC and PrPN3F4 cells cultured on glass coverslips were fixed with 4% paraformaldehyde and 0.01% glutaraldehyde and immunostained with 8H4 (1:20 dilution), followed by peroxidase-conjugated anti-mouse antibody (1:50) and mouse peroxidase antiperoxidase (1:250). After washing with PBS, the cells were exposed to 3,3'-diaminobenzidine (DAB) containing 0.1 M imidazole to obtain an electron-dense reaction product. Cells were then fixed again for 2 h at room temperature in a 0.05 M phosphate-buffered solution (pH 7.4) containing 2.5% glutaraldehyde, 2% paraformaldehyde, and 4% sucrose and postfixed in 1% osmium tetroxide for 1 h. Fixed samples were dehydrated in increasing concentrations of ethanol and embedded in Epon 812. Some ultrathin sections were stained with 2% uranyl acetate in 50% methanol and with lead citrate. Other sections were stained with lead citrate only and examined using a Zeiss CEM902 electron microscope (Oberkochen, Germany).
Surface biotinylation. Transfected PrPC or PrPN3F4 cells were washed with PBS and incubated with 1 mg/ml of sulfo-NHS-biotin in PBS for 15 min on ice. Subsequently, cells were washed with 50 mM glycine in PBS to quench excess biotin, washed three times in PBS, and lysed in NP-40 (1%) and deoxycholate (0.5%) in Tris-buffered saline (20 mM Tris, 150 mM NaCl, pH 7.4) containing 10 µg/ml each of leupeptin, antipain, and pepstatin; 1 mM phenylmethylsulfonyl fluoride; and 10 mM EDTA. Biotin-tagged proteins were retrieved from the lysate with streptavidin-coated beads, and intracellular proteins in the remaining lysate were precipitated with cold methanol. Both samples were immunoblotted with PrP-specific antibodies.
Cell homogenization and PK treatment. PrPC and PrPN3F4 cells were washed with PBS and homogenized on ice in a buffer containing 10 mM HEPES (pH 7.9), 1.5 mM MgCl2, 10 mM KCl, and 0.5 mM dithiothreitol by 20 strokes of a Kontes all-glass Dounce homogenizer. The homogenate was checked microscopically for cell breakage and centrifuged to pellet nuclei. The resulting supernatant was centrifuged at 20,000 x g to pellet membrane vesicles. These were resuspended in 0.5 ml of transport buffer (25 mM HEPES, pH 7.4, 115 mM potassium acetate, 2.5 mM MgCl2, 10 mM KCl, 2.5 mM CaCl2, and 1 mM dithiothreitol) and treated with 60 µg/ml of proteinase K (PK) on ice for 30 min. After addition of 5 mM of phenylmethylsulfonyl fluoride to stop the reaction, membrane vesicles were repelleted, solubilized in lysis buffer, and processed for Western blotting.
Immunoprecipitation and Western blotting. PrPC and PrPN3F4 cells were radiolabeled with 40 µCi/ml of Trans35S-label overnight in Dulbecco's modified Eagle's medium supplemented with 5% dialyzed serum. Labeled cells were washed with PBS, and cell lysates were subjected to immunoprecipitation with 3F4 or 8H4 and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and fluorography essentially as described previously (12). For Western blotting, transfected cells were lysed, and cold-methanol-precipitated proteins were electrophoretically transferred to Immobilon-P membranes (Millipore, Billerica, MA) for 2.5 h at 70 V and 4°C. The membrane containing transferred proteins was probed with 8H4 (1:1,000), 8B4 (1:1,000), 3F4 (1:40,000), anticalnexin (1:1,000), or streptavidin-horseradish peroxidase (1:40,000), followed by appropriate secondary antibodies conjugated to horseradish peroxidase (1:4,000). Reactive bands were visualized on an autoradiographic film by ECL (Amersham, Piscataway, NJ).
Enzymatic deglycosylation. For deglycosylation with endo-H or N-glycosidase F, half of the proteins in the sample buffer were reprecipitated with cold methanol and deglycosylated essentially as described previously (12).
TUNEL staining. Terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) staining of PrPC and PrPN3F4 cells was performed using an in situ cell death detection kit, TMR red (catalog no. 12156792910), obtained from Roche Diagnostics (Mannheim, Germany).
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The PrP constructs used in this study are depicted diagrammatically in Fig. 1B. To engineer an 3F4 epitope in the N-SP of PrP, as a first step, the internal 3F4 epitope of human PrPC was disrupted by replacing methionine at residues 109 and 112 with the corresponding mouse residues leucine and valine, respectively (PrPC(–3F4)). Subsequently, eight amino acids, including an initiating methionine and the 3F4 epitope (MAKTNMKH), were inserted before the first methionine of PrP such that residues –3 and 1 constituted the inserted 3F4 epitope (MKHM), leaving the entire N-SP sequence of PrP intact (PrPN3F4). Residues KTN were included to improve 3F4 reactivity. A construct containing two 3F4 epitopes, one at residues –3 and 1 and the other at its original site at residues 109 and 112, was also generated (PrPC/N3F4). A Flag epitope (DYKDDDDK) was engineered in the N-SP immediately following the first methionine residue of PrP (PrPNFlag). All constructs were expressed in human (M17) and mouse (N2a) neuroblastoma and CHO cells. Experiments were conducted on transiently transfected cells using three anti-PrP antibodies: 3F4, specific for N-SP in PrPN3F4 (residues –3 to 1) and additional residues 109 and 112 in PrPC/N3F4; 8B4, against N-terminal residues 35 to 45; and 8H4, against C-terminal residues 145 to 180 in PrPC and all engineered constructs. Anti-Flag antibody was used to identify PrPNFlag in transfected cells.
Substitution of the 3F4 epitope does not interfere with the biogenesis of PrPC(–3F4). To check if substitution of the 3F4 epitope in human PrPC with corresponding mouse residues alters the processing and transport of PrPC(–3F4), M17 cells expressing PrPC or PrPC(–3F4) were immunostained with 8H4, 8B4, or 3F4, followed by FITC-conjugated anti-mouse antibody. As expected, PrPC shows strong plasma membrane and intracellular reactivity with all three antibodies (Fig. 1C, panels 1 to 6). PrPC(–3F4), on the other hand, shows strong plasma membrane and intracellular reactions only with 8H4 and 8B4 and not with 3F4 antibody (Fig. 1C, panels 7 to 12). Mock-transfected M17 cells show minimal reactivity with all three antibodies on the cell surface and intracellularly (Fig. 1C, panels 13 to 18). Thus, PrPC(–3F4) shows a cellular distribution similar to that of PrPC and, as expected, does not react with 3F4 antibody. Similar results were obtained with transfected CHO and N2a cells (data not shown), confirming that PrPC(–3F4) is a suitable model for tagging the N-SP with the 3F4 epitope.
Cells expressing PrPN3F4 on the plasma membrane show abnormal cell division. Next, the effect of 3F4 epitope insertion on the biogenesis of PrPN3F4 was investigated with transfected cells. Soon it became apparent that PrPN3F4 cells died within a week of transfection, necessitating evaluation with transiently transfected cells. Surprisingly, M17 and CHO cells expressing PrPN3F4 were easily identified even without immunostaining due to their abnormally large size, aberrant shape, and atypical cell division patterns (Fig. 2).
|
Next, the plasma membranes of M17 and CHO cells expressing PrPN3F4 were immunostained with 8H4-anti-mouse antibody-FITC, followed by permeabilization with Triton X-100 and intracellular staining with antitubulin-anti-rabbit antibody-tetramethyl rhodamine isocyanate (TRITC). Nuclei were stained with Hoechst stain as described above. As expected, PrPC cells show a clear separation between dividing cells and normal intracellular distribution of tubulin (Fig. 2B, panel 1). In contrast, PrPN3F4-expressing M17 and CHO cells remain linked through tubulin filaments separated by a PrP-reactive plate (Fig. 2B, panels 2 and 3).
Ultrastructural analysis of mock-transfected and PrPC- and PrPN3F4-expressing M17 cells immunostained with 8H4 and reacted with peroxidase-conjugated secondary antibody shows a diaminobenzidine reaction on the plasma membranes of PrPC and PrPN3F4 cells but not on those of mock-transfected M17 cells (Fig. 2C, compare panels 2 to 4 with panel 1). Dividing mock-transfected and PrPC cells show distinct, complete separation of the plasma membranes, and the latter also show vesicles contributing to the generation of new membrane (Fig. 2C, panels 1 and 2). In contrast, PrPN3F4 cells are large with multiple nuclei (Fig. 2C, panel 3) or show indistinct separation of the plasma membranes of adjacent cells (Fig. 2C, panel 4). Prominent filaments are seen radiating from the intercellular membrane junction (Fig. 2C, panel 4).
Together, the above data show that PrPN3F4 cells are unable to undergo successful karyokinesis and cytokinesis and are arrested in different stages of cell division. Some of these cells show obvious pyknotic nuclei, suggesting that the PrPN3F4-induced defect in cell division is a terminal phenomenon.
Plasma membrane expression of PrPN3F4 is atypical.
To evaluate whether PrPN3F4 is linked to the plasma membrane by a GPI anchor or the transmembrane region, domain-specific anti-PrP antibodies 8H4, 8B4, and 3F4 were used for immunostaining of nonpermeabilized M17 cells. When reacted with the C-terminal antibody 8H4, a strong immunoreaction is detected, as shown in Fig. 2. However, no detectable reaction is seen with 8B4 and 3F4 antibodies on the cell surface (Fig. 3A, panels 1, 3, and 5). Although the absence of 3F4 reactivity could result from processing or inaccessibility of the N-SP, the lack of an 8B4 reaction was surprising, indicating that the N-terminal domain either is truncated or faces the cytosol and is inaccessible to added antibodies. Following permeabilization of the plasma membrane, a positive immunoreaction is detected with all three antibodies, favoring the latter possibility (Fig. 3A, panels 2, 4, and 6). Reaction with 8H4 and 8B4 is distributed at the plasma membrane, around the nucleus, and in vesicular structures (Fig. 3A, panels 2 and 4), while 3F4 reactivity is concentrated in nuclei and in aggresome-like structures (Fig. 3A, inset), with the latter often indenting the nucleus. Some cells (
10%) show 3F4 reactivity in vesicular structures in the cytosol (Fig. 3A, panel 6).
|
PrPN3F4 forms with cleaved and intact N-SP are present on the plasma membrane. Differential staining of PrPN3F4 with domain-specific antibodies suggested the presence of N-terminally truncated or full-length C-transmembrane forms on the plasma membranes of M17 and CHO cells. To distinguish between these possibilities, plasma membrane proteins of mock-transfected, PrPC, and PrPN3F4 cells were biotinylated with sulfo-NHS-biotin, and biotin-tagged proteins were retrieved from cell lysates with streptavidin-coated beads. Unbound intracellular proteins were precipitated with cold methanol, and both surface and intracellular samples were fractionated by SDS-PAGE and immunoblotted with 8H4 and 3F4.
Surface proteins from mock-transfected M17 and CHO cells show barely any reaction with 8H4 and 3F4 antibodies (Fig. 4A, lanes 1 and 8), while PrPC cells show the expected glycoforms of PrPC migrating at 27, 30 to 32, and 35 to 38 kDa when immunoblotted with 8H4 (Fig. 4A, lane 2). Surprisingly, proteins retrieved from the plasma membrane of PrPN3F4 M17 cells show a strong 8H4-reactive 29-kDa band, a less intense 27-kDa band, and slower-migrating forms of 30 to 32 and 35 to 37 kDa (Fig. 4A, lane 4), while the intracellular sample shows indistinct bands, some of which comigrate with surface-expressed PrPN3F4 forms (Fig. 4A, lane 3). Evaluation of CHO cells in parallel shows similar intracellular and surface bands (Fig. 4A, lanes 5 to 8), except that the 27-kDa band is absent from the surface (Fig. 4A, lane 6). The 29-kDa band reacts with 3F4, indicating that it includes the uncleaved N-SP of PrPN3F4 (Fig. 4A, lane 7). Mock-transfected cells do not show any 3F4-reactive forms on the cell surface (Fig. 4A, lane 8).
|
PrPN3F4 is inserted in the C-transmembrane orientation on the plasma membrane. A positive immunoreaction with only 8H4 (Fig. 2A and 3A), despite the presence of full-length PrPN3F4 forms with uncleaved N-SP on the cell surface (Fig. 4A), indicated a C-transmembrane orientation of PrPN3F4 on the plasma membranes of M17 and CHO cells. In this orientation, only the C-terminal 8H4 epitope would be accessible to added antibodies, since 8B4 and 3F4 epitopes face the cytosol (Fig. 1A). Thus, sequential staining of nonpermeabilized cells with C-terminal antibody 8H4 (green) followed by N-terminal antibody 8B4 (red) would reveal GPI-linked full-length (orange), C-transmembrane and N-terminally truncated (green), and N-transmembrane and C-terminally truncated (red) PrPN3F4 forms on the plasma membrane (Fig. 4B, schematic). If the immunostaining pattern changes from 8H4 only (green) in nonpermeabilized cells to 8H4 and 8B4 (orange) in cells immunostained with 8H4, permeabilized, and subsequently reacted with 8B4, it would indicate the presence of CtmPrP on the plasma membrane (Fig. 4C, schematic). In addition, this method would allow a rough estimation of the full-length and truncated PrPN3F4 forms on the cell surface and intracellularly. Accordingly, PrPN3F4-expressing M17 cells were incubated with 8H4-anti-mouse antibody-FITC followed by excess unlabeled anti-mouse immunoglobulin to block unoccupied 8H4 binding sites. Subsequently, the same cells were immunostained with 8B4-anti-mouse antibody-TRITC. In nonpermeabilized PrPC cells, 8H4 and 8B4 staining colocalizes in most areas, as is expected of GPI-linked PrPC, where both epitopes would be on the exoplasmic face of the plasma membrane (Fig. 4B, panels 1 to 3). Some reactivity is restricted to 8H4, which is indicative of N-terminally truncated forms such as the 18-, 20-, and 24-kDa fragments reported earlier (Fig. 4B, panels 1 to 3) (12). In contrast, PrPN3F4 cells show strong a reaction with 8H4 but no reactivity with 8B4 on the plasma membrane (Fig. 4B, panels 4 to 6), indicating the presence of CtmPrPN3F4 or N-terminally truncated PrPN3F4 on the plasma membrane. Unexpectedly, strong 8B4 reactivity is detected in localized regions near the membrane, probably representing N-terminal fragments that are being secreted (Fig. 4B, panels 4 to 6).
Similar immunoreaction of permeabilized PrPN3F4 cells is depicted in Fig. 4C. The images in panels 1 to 3 focus on two cells at different depths, one near the surface and the other near the center of the cell. These cells are enlarged in panels 1a to 3a (surface) and 1b to 3b (intracellular). Sequential immunoreaction with 8H4-anti-mouse antibody-FITC, unlabeled anti-mouse immunoglobulin G, and 8B4-anti-mouse antibody-TRITC shows colocalization of 8H4 and 8B4 at the plasma membrane (Fig. 4C, panels 1a to 3a), unlike the staining pattern observed without permeabilization (compare Fig. 4B and C). Additional areas of colocalization include the perinuclear region and cytoplasmic vesicles (Fig. 4C, panels 1b to 3b). In certain regions 8H4 as well as 8B4 reactivity is detected independent of the other, indicating the presence of N- and C-terminally truncated fragments in the same cell (Fig. 4C, panels 1a and b to 3a and b). These results confirm the presence of full-length and truncated CtmPrPN3F4 forms on the plasma membrane and in intracellular compartments. Immunostaining with 8H4-FITC followed by 3F4-TRITC shows colocalization in aggresomes, indicating the presence of untranslocated, N-SP-containing full-length PrPN3F4 forms in that locale. The additional presence of 8H4 and 3F4 reactivity in vesicular structures is suggestive of translocated forms on their way to the plasma membrane (Fig. 4C, panels 4 to 6). The aggresomes are large, perinuclear, and surrounded by a vimentin cage (Fig. 4C, panels 7 to 9), whereas vesicular structures are small and punctuate and are distributed in the cytosol. Lack of 3F4 reactivity at the plasma membrane probably reflects inaccessibility of this epitope following permeabilization of fixed membrane. As is typical of aggresomes, PrPN3F4-containing aggregates are enclosed by a vimentin cage (Fig. 4C, panels 7 to 9). As noted in Fig. 3A, panel 6, some cells show nuclear localization of 3F4 reactivity (Fig. 4C, panels 7 to 9).
Finally, the Ctm orientation of PrPN3F4 was confirmed by subjecting microsomes isolated from PrPC and PrPN3F4 cell homogenates to limited digestion with proteinase K in the absence of detergent. Under the conditions used, 99% of Golgi complex- and ER-derived microsomes reseal in the correct orientation and enclose PrPC within membrane vesicles, while plasma membrane sheets do not reseal, exposing surface PrP forms to added PK. The exposed domains of transmembrane forms of PrP are proteolyzed by PK, sparing a 19- to 20-kDa C-terminal fragment of CtmPrP and a 14-kDa N-terminal fragment of NtmPrP, respectively (16) (Fig. 1A). The internal 3F4 epitope at residues 109 and 112 of PrPC is protected by the lipid bilayer in both transmembrane forms, rendering this antibody an ideal tool for the detection of transmembrane PrP forms. However, since PrPN3F4 lacks the internal 3F4 epitope, an additional PrP form where the internal 3F4 epitope has been reintroduced in PrPN3F4 (PrPC/N3F4) was used to confirm the identity of PK-protected fragments from PrPN3F4. In the experiment described below, 8H4 was used to identify PK-generated fragments of PrPN3F4, and 8H4 and 3F4 were used for PrPC and PrPC/N3F4. The samples were not deglycosylated to facilitate easy differentiation between the 18-kDa C-terminal fragment of PrPC and a similar 19- to 20-kDa fragment of CtmPrP. The former is a product of normal recycling from the plasma membrane and lacks the internal 3F4 epitope (6), while the latter is a PK-protected fragment of CtmPrP and includes the internal 3F4 epitope.
In the experiment described below, PrPC and PrPN3F4 microsomes were treated with PK on ice, and the lysates were fractionated by SDS-PAGE and analyzed by Western blotting. Immunoblotting of untreated PrPC lysates with 3F4 reveals the expected glycoforms migrating at 27, 30 to 32, and 35 kDa and a minor fragment of 20 kDa (band 1) (12). Following PK treatment, full-length PrPC forms are partially cleaved, the intensity of the 20-kDa band increases (band 1), and a new fragment of 14 kDa (band 2) is detected with 3F4 (Fig. 5A, lanes 1 and 2). Immunoblotting of the same membrane with 8H4 reveals normal PrPC glycoforms and 24- and 18-kDa C-terminal fragments in the untreated samples (Fig. 5A, lane 3) (6). Following PK treatment, the intensities of the 34- to 35-kDa full-length forms decrease and those of the 24-, 20-, and 18-kDa forms increase significantly (Fig. 5A, lane 4). The 20-kDa form shows slight but certain reactivity with 8H4, but the 14-kDa fragment is not recognized by this antibody (Fig. 5A, lane 4). Thus, the 20-kDa fragment is C terminal and 14-kDa fragment N terminal, indicating their origins from CtmPrP and NtmPrP, respectively. The increased intensities of 24- and 18-kDa fragments following PK treatment indicate cleavage of PrPC forms exposed on plasma membrane sheets at the normal cleavage site at residues 111/112 (Fig. 5A, lane 4) (6). Microsomes of PrPN3F4 cells processed in parallel and immunoblotted with 8H4 due to the absence of 3F4 epitope at residues 109 to 112 show PrPN3F4 glycoforms of 27 to 35 kDa and fragments of 24, 20, and 19 kDa (Fig. 5A, lane 5; also see Fig. 6). PK treatment influences almost all PrPN3F4 forms except the 32-kDa band. The 27-, 29-, and 35-kDa forms are decreased, whereas the 24-, 20 (band 1)-, 19 (band 3)-, and 12 (band 4)-kDa forms are significantly increased (Fig. 5A, lanes 5 and 6). Subsequent immunoblotting of the same membrane with 8B4 shows reactivity with 27-, 29-, and 32-kDa forms but not with fragments of 24, 20, 19, and 12 kDa, indicating their origin from the C terminus (Fig. 5A, lane 7). Harsh treatment with PK digests all PrPN3F4 forms, leaving a bold 32-kDa band and minor bands of 27, 19, and 15 kDa (Fig. 5A, lane 8). (Although the migrations of the 18-kDa fragment of PrPC and the 19-kDa fragment of PrPN3F4 are very similar, the latter includes the internal 3F4 epitope [Fig. 5B] and thus must contain additional amino acid residues.)
|
|
Together with the immunofluorescence data, the above results show that at steady state, all PrPN3F4 molecules that translocate into the ER are inserted in the Ctm orientation in the lipid bilayer. Of these, 40% are truncated proximal to residue 109 (sparing the internal 3F4 epitope) to generate C-terminal fragments of 19 and 20 kDa. Notably, both full-length PrPN3F4 and C-terminal fragments are transported to the cell surface and expressed on the plasma membrane of M17 cells in the Ctm orientation.
Biosynthesis and processing of CtmPrPN3F4 in M17 cells. The synthesis and processing of PrPN3F4 at early time points were evaluated by pulse-chase analysis. Accordingly, PrPC and PrPN3F4 cells were pulsed with [35S]methionine for 30 min and chased for 0, 1, 2, and 4 h. At the indicated times, cell lysates were immunoprecipitated with 3F4 or 8H4, and eluted proteins were analyzed by SDS-PAGE and fluorography. Immediately following the pulse, three glycoforms of PrPC are detected with 3F4 and 8H4: the unglycosylated (band 1), monoglycosylated (band 2), and diglycosylated (band 3) forms containing high-mannose glycans (Fig. 6A, lanes 1 and 5). With increasing chase times, the immature glycans of mono- and diglycosylated forms are modified, resulting in the mature glycosylated form (band 5) (Fig. 6A, lanes 2 to 4 and 6 to 8). Similar analysis of PrPN3F4 lysates with 3F4 reveals bands that migrate 2 kDa slower than corresponding PrPC bands (bands 1+ to 3+) and an additional band that migrates at 32 kDa (band 4++). The slower migration of these forms is due to the presence of uncleaved N-SP in bands 1+, 2+, and 3+ and additional uncleaved GPI-SP in band 4++ (Fig. 6A, lane 9). Following a chase of 1 h, most 3F4-reactive PrPN3F4 forms are lost. Only minor amounts of band 3+ remain (Fig. 6A, lanes 9 to 12). Immunoprecipitation with 8H4 reveals an additional band at 27 kDa at time zero (band 1), indicating cleavage of N- and GPI-SP and mature bands similar to PrPC following the chase. However, the overall intensity of mature PrPN3F4 bands is significantly less than that of PrPC (Fig. 6A, lanes 13 to 16).
In order to visualize minor forms of PrPC and PrPN3F4 generated at early time points following synthesis, twice the number of PrPN3F4 cells were pulsed with [35S]methionine for 30 min, chased for 1 hour, and processed for immunoprecipitation as described above. PrPC lysates from both the pulse and chase samples immunoprecipitated with 3F4 and 8H4 reveal immature glycoforms ranging from 27 to 32 kDa and a 20-kDa band representing the N-terminally truncated Ctm form that is generated in the ER (Fig. 6B, lanes 1 to 4) (12). Immunoprecipitation of PrPN3F4 lysates with 3F4 reveals both immature glycoforms migrating 2 kDa slower than corresponding PrPC forms and fragments of 24 and 19 kDa (Fig. 6B, lane 5). Notably, 3F4-reactive and hence N-SP-containing mature forms are detected following 1 hour of chase, while the fragments turn over (Fig. 6B, lane 6). Immunoprecipitation with 8H4 shows full-length forms similar to the glycoforms observed in lane 5 and corresponding N-SP-cleaved forms 1, 2, and 3 (Fig. 6B, lanes 7 and 8). In addition, fragments of 24, 20, and 19 kDa that do not react with 3F4 are detected (Fig. 6B, lane 7). The 24-kDa fragment undergoes further processing, and the 19-kDa fragment increases in intensity with chase (Fig. 6B, lanes 7 and 8).
Together, these results highlight certain key aspects of PrPN3F4 synthesis and processing. (i) The N-SP of the majority of PrPN3F4 is cleaved within 1 hour of synthesis, although the rate of cleavage is much slower than that of PrPC. This is surprising since a significant amount of PrPN3F4 on the plasma membrane contains uncleaved N-SP, indicating that this form accumulates at steady state. (ii) The uncleaved N-SP does not interfere with N glycosylation, since all N-SP-containing PrPN3F4 forms except bands 1+ and 4++ acquire immature glycans. A limited number are further processed and contain mature glycans. (iii) The C-terminal 19-kDa fragment is generated within 30 min of synthesis and increases with chase. (iv) Form 4++, constituting 5% of synthesized PrPN3F4, may represent untranslocated forms in the cytosol with intact N-SP and GPI-SP. (v) The unglycosylated N-SP-cleaved PrPN3F4 form migrates a little faster than the corresponding PrPC form due to a posttranslational event that is not clear from our data.
Transport and processing of CtmPrP are cell type and N-SP sequence specific. To resolve the confounding presence of CtmPrPN3F4 on the plasma membranes of human neuroblastoma (M17) and CHO cells and its complete absence from the plasma membrane of mouse neuroblastoma (N2a) cells, steady-state expression of PrPC and PrPN3F4 was evaluated in N2a cells by immunoblotting (Fig. 7A). Reaction with 3F4 reveals the expected glycoforms of PrPC that are resistant to digestion with endo-H, indicating the presence of Golgi complex-modified glycans. Deglycosylation reveals the full-length 27-kDa form and an additional 20-kDa fragment (Fig. 7A, lanes 1 to 3). Antibodies 8B4 and 8H4 detect similar PrPC glycoforms, except that 8H4 reveals an additional C-terminal fragment of 18 kDa (Fig. 7A, lanes 7 to 9 and 13 to 15, respectively). Similar evaluation of PrPN3F4 shows a 32-kDa form that acquires immature glycans (Fig. 7A, lanes 4 to 6, 10 to 12, and 16 to 18). Minor proteolytically processed forms are also detected with all three antibodies.
|
These results demonstrate that the processing of CtmPrPN3F4 differs significantly in human and mouse neuroblastoma cells. In human cells CtmPrPN3F4 is processed and transported to the plasma membrane, whereas in mouse cells it is retained in the ER. PrPNFlag is not transported from the ER in all three cell lines.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When expressed in human neuroblastoma and CHO cells, 80% of CtmPrPN3F4 undergoes proteolytic cleavage, probably at the ER membrane, releasing the N-terminal region in the cytosol. The cleaved C-terminal domain and 20% of full-length CtmPrPN3F4 molecules are transported to the plasma membrane. In contrast, 100% of PrPN3F4 in mouse neuroblastoma cells is retained in the ER. PrPNFlag is not transported from the ER in all three cell lines examined. It is interesting to note that only human neuroblastoma and CHO cells show atypical karyokinesis and cytokinesis, although PrPN3F4 and PrPNFlag are also in the Ctm orientation in all three cell lines (see below). A variety of defects in cytokinesis and karyokinesis are noted; the majority of the cells (60%) contain multiple nuclei and are connected to smaller cells that contain pyknotic nuclei, some cells (18%) remain connected by a long extension of the plasma membrane and fail to separate, while others (20%) are abnormally large and contain multiple nuclei without complete separation of the nuclear membrane or any obvious attempt at cytokinesis. In this regard, it is interesting to note that abnormal induction of cell cycle proteins in adult neurons is the proximal cause of neuronal death in several neurodegenerative disorders (2). Although upregulation of cell cycle markers is difficult to demonstrate in culture-adapted cell lines, our data clearly implicate dysregulation of the cell cycle in CtmPrP-mediated neuronal death, adding prion disorders to the growing list of diseases where neuronal cell division triggers neuronal death.
How might CtmPrPN3F4 induce abnormal cell division? We propose a combination of cis- and trans-acting factors in triggering this process. In support of cis-acting factors, the transmembrane domain of PrP is known to destabilize the membrane, and a fragment representing this domain confers fusogenic properties and induces apoptosis in cortical neurons in vitro and in retinal neurons in vivo (4). Such an effect of CtmPrPN3F4 at the plasma membrane could interfere with cytokinesis, a defect consistently observed in cells expressing this form on the cell surface. The cis-acting effect can be attributed to the uncleaved N-SP of CtmPrPN3F4, as demonstrated for untranslocated PrP molecules in the cytosol (19). It is unclear from our data whether the uncleaved N-SP of CtmPrPN3F4 acts as an additional membrane anchor or merely interacts with the lipid bilayer. Given its hydrophobic nature, either or both of these interactions are likely. Our inability to immunostain the N-SP of CtmPrPN3F4 at the plasma membrane despite its certain presence in that locale suggests its interaction with the lipid bilayer, thereby obscuring the 3F4 epitope. The alternate possibility that the 3F4 epitope in the N-SP is less reactive is unlikely, since strong reactivity with PrPN3F4 sequestered in aggresomes is observed using the same antibody. Thus, membrane perturbation by the transmembrane domain and the N-SP may induce defective cytokinesis.
The presence of trans-acting factors is evident from the deleterious effect of CtmPrPN3F4 on nuclear division. Either certain soluble factors or the N-terminal domain of truncated CtmPrPN3F4 alters nuclear division. The N terminus of CtmPrP contains nuclear localization signals (14) and is released in the cytosol due to its cleavage at the ER membrane (12, 13; this report). We could detect this fragment within the nucleus in only 10% of the cells, perhaps due to its rapid degradation by nuclear and cytosolic proteasomes (14). Nevertheless, under conditions where proteasome function is compromised, this fragment may accumulate in the nuclei and alter transcription or interfere with chromosomal segregation, resulting in failed karyokinesis as observed in CtmPrPN3F4-expressing cells. An interaction of PrP with DNA and accumulation of PrPSc in the nuclei of infected cells support this assumption (9, 25). Several other proteins have been reported to influence cellular function through regulated proteolysis followed by translocation of the truncated fragment to the nucleus, such as the C-terminal fragment of APP (5), the N-terminal domain of mammalian transcription factor ATF6 (15), and the sterol regulatory element binding proteins (3). Signal peptide-derived fragments also serve as trans-acting effectors (26). It is therefore likely that a similar phenomenon occurs in CtmPrPN3F4-expressing cells. The above scenario would explain the absence of abnormal cell division in cells where CtmPrP does not undergo N-terminal cleavage, as noticed for CtmPrPN3F4 in mouse neuroblastoma cells and for CtmPrPNFlag in all three cell lines tested. The Ctm orientation of PrPN3F4 in mouse neuroblastoma cells and of PrPNFlag in all three cell lines is evident from the presence of intact N-SP and immature glycans, indicating a cytosolic location of the N-SP and ER localization of the C terminus (31). Yet these CtmPrP forms do not affect cell division, supporting our contention that N-terminal truncation and transport of CtmPrP to the plasma membrane are necessary to induce abnormal cell division.
Surprisingly, some cells remain connected long after cell division and undergo apoptosis without apparent dissolution of the postmitotic bridge. A similar defect in cytokinesis has been observed in cells that express the mammalian polo-like kinase 3, lack annexin 11, or are depleted of centriolin, syntaxin-2, and endobrevin, to mention a few (7, 11, 23, 35). Whether any of these conditions are induced by CtmPrPN3F4 is unclear from our data, but this raises the intriguing possibility that CtmPrP modulates cell cycle proteins through complex signaling pathways.
Factors influencing the biogenesis and transport of CtmPrP are equally intriguing. In human neuroblastoma and CHO cells, PrPN3F4 exists as full-length forms with and without the N-SP and N-terminally truncated forms. The N-SP influences PrP membrane orientation and its own cleavage by modulating the gating properties of the translocon, allowing for the synthesis of CtmPrP with intact or cleaved N-SP (22, 30). A similar cleavage of the N-SP within the translocon has been reported for other polypeptides that fail to enter the ER lumen (10, 27). The N-terminal fragments of CtmPrP are detected in the nucleus and cytosol, often concentrated in cellular processes in the latter. In mouse neuroblastoma cells, neither the N-SP nor the GPI-SP of PrPN3F4 is removed, based on mobility on SDS-PAGE and reactivity with 3F4, but the protein is translocated into the ER, as evidenced by acquisition of endo-H-sensitive glycans. These results clarify the conflicting reports regarding the cellular localization and toxic potential of CtmPrP in different cell lines and mouse models (16-18, 31-34).
In conclusion, our data demonstrate the overriding influence of cis- and trans-acting factors on the biogenesis and toxic potential of CtmPrP. It appears that two specific events modulate the synthesis and toxicity of CtmPrP. First, interaction of specific residues in the N-SP of PrP, ER translocon proteins, and trans-acting factors that may be species or cell specific determine the percentage of PrP that is translocated fully into the ER lumen and is thus GPI linked, assumes a Ctm or Ntm orientation, or is not translocated at all. Second, efficiency of the cellular quality control determines whether the untranslocated, translocated, and transmembrane forms undergo degradation or are processed further and transported to the plasma membrane. CtmPrP forms that escape degradation interfere with cytokinesis, and the N-terminal fragment of CtmPrP possibly interacts with nuclear chromatin, interfering with karyokinesis. The fact that N-SP sequences play a direct role in CtmPrP transport and cytotoxicity may have significant implications for familial prion disorders with mutations in the GPI-SP of PrP, since this peptide has been reported to function as an inefficient signal for ER translocation in vitro (20). If such a situation occurs in vivo, mutations in this region may predispose to the generation of CtmPrP with consequent cytotoxicity. However, no mutations in the N-SP of PrP have been reported so far, and the specific factors influencing PrP topology are unclear. Further investigations are needed to resolve these perplexing issues. Although quantification of CtmPrP in prion-affected humans and mouse models remains a technical challenge, this study provides insight into the mechanisms by which it induces neuronal death.
|
ACKNOWLEDGMENTS |
|---|
This study was supported by NIH grants NS39089 and NS35962 to N.S.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. Becker, E. B., and A. Bonni. 2004. Cell cycle regulation of neuronal apoptosis in development and disease. Prog. Neurobiol. 72:1-25.[CrossRef][Medline]
3. Brown, M. S., and J. L. Goldstein. 1997. The SREBP pathway: regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell 89:331-340.[CrossRef][Medline]
4. Chabry, J., C. Ratsimanohatra, I. Sponne, P. P. Elena, J. P. Vincent, and T. Pillot. 2003. In vivo and in vitro neurotoxicity of the human prion protein (PrP) fragment P118-135 independently of PrP expression. J. Neurosci. 23:462-469.
5. Chen, Y., W. Liu, D. L. McPhie, L. Hassinger, and R. L. Neve. 2003. APP-BP1 mediates APP-induced apoptosis and DNA synthesis and is increased in Alzheimer's disease brain. J. Cell Biol. 163:27-33.
6. Chen, S. G., D. B. Teplow, P. Parchi, J. K. Teller, P. Gambetti, and L. Autilio-Gambetti. 1995. Truncated forms of the human prion protein in normal brain and in prion diseases. J. Biol. Chem. 270:19173-19180.
7. Conn, C. W., R. F. Hennigan, W. Dai, Y. Sanchez, and P. J. Stambrook. 2000. Incomplete cytokinesis and induction of apoptosis by overexpression of the mammalian Polo-like kinase, Plk3. Cancer Res. 60:6826-6831.
8. Dimcheff, D. E., J. L. Portis, and B. Caughey. 2003. Prion proteins meet protein quality control. Trends Cell Biol. 13:337-340.[CrossRef][Medline]
9. Gabus, C., S. Auxilien, C. Pechoux, D. Dormont, W. Swietnicki, M. Morillas, W. Surewicz, P. Nandi, and J. L. Darlix. 2001. The prion protein has DNA strand transfer properties similar to retroviral nucleocapsid protein. J. Mol. Biol. 307:1011-1021.[CrossRef][Medline]
10. Garcia, P. D., J. H. Ou, W. J. Rutter, and P. Walter. 1988. Targeting of the hepatitis B virus precore protein to the endoplasmic reticulum membrane: after signal peptide cleavage translocation can be aborted and the product released into the cytoplasm. J. Cell Biol. 106:1093-1104.
11. Gromley, A., A. Jurczyk, J. Sillibourne, E. Halilovic, M. Mogensen, I. Groisman, M. Blomberg, and S. Doxsey. 2003. A novel human protein of the maternal centriole is required for the final stages of cytokinesis and entry into S phase. J. Cell Biol. 161:535-545.
12. Gu, Y., H. Fujioka, R. S. Mishra, R. Li, and N. Singh. 2002. Prion peptide 106-126 modulates the aggregation of cellular prion protein and induces the synthesis of potentially neurotoxic transmembrane PrP. J. Biol. Chem. 277:2275-2286.
13. Gu, Y., S. Verghese, R. S. Mishra, X. Xu, Y. Shi, and N. Singh. 2003. Mutant prion protein-mediated aggregation of normal prion protein in the endoplasmic reticulum: implications for prion propagation and neurotoxicity. J. Neurochem. 84:10-22.[CrossRef][Medline]
14. Gu, Y., J. Hinnerwisch, R. Fredricks, S. Kalepu, R. S. Mishra, and N. Singh. 2003. Identification of cryptic nuclear localization signals in the prion protein. Neurobiol. Dis. 12:133-149.[CrossRef][Medline]
15. Haze, K., H. Yoshida, H. Yanagi, T. Yura, and K. Mori. 1999. Mammalian transcription factor ATF is synthesized as a transmembrane protein and activated by proteolysis in response to endoplasmic reticulum stress. Mol. Biol. Cell 10:3787-3799.
16. Hegde, R. S., J. A. Mastrianni, M. R. Scott, K. A. DeFea, P. Tremblay, M. Torchia, S. J. DeArmond, S. B. Prusiner, and V. R. Lingappa. 1998. Transmembrane form of the prion protein in neurodegenerative disease. Science 279:827-834.
17. Hegde, R. S., S. Voigt, and V. R. Lingappa. 1998. Regulation of protein topology by trans-acting factors at the endoplasmic reticulum. Mol. Cell 2:85-91.[CrossRef][Medline]
18. Hegde, R. S., P. Tremblay, D. Groth, S. J. DeArmond, S. B. Prusiner, and V. R. Lingappa. 1999. Transmissible and genetic prion diseases share a common pathway of neurodegeneration. Nature 402:822-826.[CrossRef][Medline]
19. Heller, U., K. F. Winklhofer, J. Heske, A. Reintjes, and J. Tatzelt. 2003. Post-translational import of the prion protein into the endoplasmic reticulum interferes with cell viability: a critical role for the putative transmembrane domain. J. Biol. Chem. 278:36139-36147.
20. Holscher, C., U. C. Bach, and B. Dobberstein. 2001. Prion protein contains a second endoplasmic reticulum targeting signal sequence located at its C terminus. J. Biol. Chem. 276:13388-13394.
21. Kim, S. J., D. Mitra, J. R. Salerno, and R. S. Hegde. 2002. Signal sequences control gating of the protein translocation channel in a substrate-specific manner. Dev. Cell 2:207-217.[CrossRef][Medline]
22. Kim, S. J., R. Rahbar, and R. S. Hegde. 2001. Combinatorial control of prion protein biogenesis by the signal sequence and transmembrane domain. J. Biol. Chem. 276:26132-26140.
23. Low, S. H., X. Li, M. Miura, N. Kudo, B. Quinones, and T. Weimbs. 2003. Syntaxin 2 and endobrevin are required for the terminal step of cytokinesis in mammalian cells. Dev. Cell 4:753-759.[CrossRef][Medline]
24. Ma, J., R. Wollmann, and S. Lindquist. 2002. Neurotoxicity and neurodegeneration when PrP accumulates in the cytosol. Science 298:1781-1785.
25. Mange, A., C. Crozet, S. Lehmann, and F. Beranger. 2004. Scrapie-like prion protein is translocated to the nuclei of infected cells independently of proteasome inhibition and interacts with chromatin. J. Cell Sci. 117:2411-2416.
26. Martoglio, B., and B. Dobberstein. 1998. Signal sequences: more than just greasy peptides. Trends Cell Biol. 8:410-415.[CrossRef][Medline]
27. Ooi, C. E., and J. Weiss. 1992. Bidirectional movement of a nascent polypeptide across microsomal membranes reveals requirements for vectorial translocation of proteins. Cell 71:87-96.[CrossRef][Medline]
28. Ott, C. M., and V. R. Lingappa. 2004. Signal sequences influence membrane integration of the prion protein. Biochemistry 43:11973-11982.[CrossRef][Medline]
29. Prusiner, S. B. 1998. Prions. Proc. Natl. Acad. Sci. USA 95:13363-13383.
30. Rane, N. S., J. Yonkovich, and R. S. Hegde. 2004. Protection from cytosolic prion protein toxicity by modulation of prion translocation. EMBO J. 23:4550-4559.[CrossRef][Medline]
31. Stewart, R. S., B. Drisaldi, and D. A. Harris. 2001. A transmembrane form of the prion protein contains an uncleaved signal peptide and is retained in the endoplasmic reticulum. Mol. Biol. Cell 12:881-889.
32. Stewart, R. S., and D. A. Harris. 2003. Mutational analysis of topological determinants in prion protein (PrP) and measurement of transmembrane and cytosolic PrP during prion infection. J. Biol. Chem. 278:45960-45968.
33. Stewart, R. S., P. Piccardo, B. Ghetti, and D. A. Harris. 2005. Neurodegenerative illness in transgenic mice expressing a transmembrane form of the prion protein. J. Neurosci. 25:3469-3477.
34. Stewart, R. S., and D. A. Harris. 2005. A transmembrane form of the prion protein is localized in the Golgi apparatus of neurons. J. Biol. Chem. 280:15855-15864.
35. Tomas, A., C. Futter, and S. E. Moss. 2004. Annexin 11 is required for midbody formation and completion of the terminal phase of cytokinesis. J. Cell Biol. 165:813-822.
This article has been cited by other articles:
| ||||||||||||
