On the same cell type GPI-anchored normal cellular prion and DAF protein exhibit different biological properties

 Anorganische Chemie

of 6

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
On the same cell type GPI-anchored normal cellular prion and DAF protein exhibit different biological properties
  On the same cell type GPI-anchored normal cellular prionand DAF protein exhibit different biological properties Ruliang Li, a Tong Liu, b Fukuoka Yoshihiro, b Magdalena Tary-Lehmann, b Mark Obrenovich, b Haydar Kuekrek, b Shin-Chung Kang, b Tao Pan, b Boon-Seng Wong, b M. Edward Medof, b and Man-Sun Sy a,b,* a School of Life Science, Wuhan University, Wuhan 430072, China b Institute of Pathology, School of Medicine, Case Western Reserve University, Room 933, BRB, Cleveland, OH 44107-1712, USA Received 20 February 2003 Abstract Normal cellular prion protein (PrP C ) and decay-accelerating factor (DAF) are glycoproteins linked to the cell surface by gly-cosylphosphatidylinositol (GPI) anchors. Both PrP C and DAF reside in detergent insoluble complex that can be isolated fromhuman peripheral blood mononuclear cells. However, these two GPI-anchored proteins possess different cell biological properties.The GPI anchor of DAF is markedly more sensitive to cleavage by phosphatidylinositol-specific phospholipase C (PI-PLC) thanthat of PrP C . Conversely, PrP C has a shorter cell surface half-life than DAF, possibly due to the fact that PrP C but not DAF is shedfrom the cell surface. This is the first demonstration that on the surface of the same cell type two GPI-anchored proteins differ intheir cell biological properties.   2003 Elsevier Science (USA). All rights reserved. The cellular prion protein (PrP C ) is a highly conservedglycoprotein linked to cell surfaces by a glycosylphos-phatidylinositol (GPI) anchor [1–3]. Abnormalities of PrP C underlie a group of rare neurodegenerative diseasescollectively known as transmissible spongiform enceph-alopathies or prion diseases [4–7]. According to currentconcepts, the pathogenic mechanism responsible for allforms of prion diseases involves the conversion of thenormal PrP C into the pathogenic scrapie prion, PrP Sc [4,8].Despite the large body of information on the impor-tance of PrP Sc in disease, the normal physiologicalfunctions of PrP C are not completely understood [4,8].Several cell surfaces as well as cytoplasmic proteins havebeen reported to interact with PrP C [9–15]. In addition,PrP C also binds nucleic acids [16,17] and glycosamino-glycans [18,19], takes up divalent cations [20,21], andfunctions as a metal transporter on the cell surface[22,23].Morerecently, ithas beenreported that PrP C hasenzymatic activity similar to that of super-oxide dismu-tase [24], and may be involved in cellular stress responsesaswellasinapoptosisofneuronalcells[25,26].Atpresenttime however, it is not clear whether PrP C functionsprincipally as a cell-surface receptor, a cytoplasmicprotein, a nuclear protein or all of the above.GPI-anchor structures are composed of a glycan inglycoside linkage to phosphatidylinositol (PI). The gly-can core consists of glucoamine, three mannose residues,and a terminal ethanolamine, which is amide bonded tothe C-terminus of the protein. GPI modification occurspost-translationally in the endoplasmic reticulum wherethe GPI modification signal sequence directs the re-moval of a signal peptide. The pre-assembled GPI an-chor is then covalently attached  en bloc  to the cleavedprotein [27]. There is evidence to suggest that the GPIanchor is important in the conversion of PrP C to PrP Sc .In vitro in a cell model, treatment of cells with phos-pholipase C (PLC), an enzyme that hydrolyzes the GPIanchor, inhibits the production of PrP Sc [28]. In anothercell line, chimeric PrP molecules with a transmembraneanchor rather than a GPI anchor are resistant to PrP Sc Biochemical and Biophysical Research Communications 303 (2003) 446–451www.elsevier.com/locate/ybbrc BBRC * Corresponding author. Fax: 1-216-368-1357. E-mail address:  MXS92@po.cwru.edu (M.-S. Sy).0006-291X/03/$ - see front matter    2003 Elsevier Science (USA). All rights reserved.doi:10.1016/S0006-291X(03)00354-1  conversion [29]. How the GPI is involved in the con-version of PrP C to PrP Sc is unknown. One possibility isthat because GPI-anchored proteins occupy micro-domains on the cell membrane termed lipid rafts [30,31],aggregation of PrP C in rafts could facilitate the inter-action between PrP C and PrP Sc [32,33]. In work done todate, however, only the GPI anchors of mouse andhamster brain derived PrP C and PrP Sc have been studiedin some detail [2,3,34].Not all GPI anchors are the same. The chemicalcomposition of the GPI anchor is cell type-specific[35,36]. The glycan core of the GPI anchor can be deco-rated with different carbohydrates, thereby creating het-erogeneity. The GPI anchor of hamster and humanbrain-derivedPrP C containssialicacid,afeaturethathasnot been found on the anchors of other GPI-anchoredproteins [34,37]. GPI-anchored proteins also can be an-choredtocellmembraneseitherbytwolipidchains(two-footedanchor)orthreelipidchains(three-footedanchor)[36,38,39].Thelengthandthechemicalnatureofthelipid(alkyl/acyl) chain can also vary. More recently, func-tionally different GPI-anchored proteins have beenshown to occupy different micro-domains on the neuro-nalcellsurface[40].Thereasonsforthisdifferencearenotknown. A better understanding of the properties of theGPI anchor of human PrP C potentially could providenew information relevant to the pathogenesis of priondiseases.Most studies of human prion protein have been car-ried out on brain tissues or tumor cell lines transfectedwith PrP C expression constructs [41]. Relatively little isknown about the expression and function of humanPrP C in normal cells. Using a panel of newly developedmonoclonal antibodies specific for human PrP C , we re-ported earlier that high levels of PrP C are present onhuman peripheral blood leukocytes including: T cells, Bcells, monocytes, and dendritic cells [42]. In this manu-script we describe studies comparing the properties of human PrP C with those of another GPI-anchored pro-tein, decay-accelerating factor (DAF) on human pe-ripheral blood leukocytes, on a human T cell clone, anda human leukemia cell line, Jurkat. We found that whileboth human PrP C and DAF are distributed principallyin rafts on the same cell type, they differ significantly intheir sensitivity to digestion with PI-PLC as well as theirturnover rate. These results indicate that prion proteinand DAF, a prototypic GPI-anchored protein that hasbeen well characterized, differ in their cell biologicalproperties. Materials and methods Monoclonal antibodies and isolation of human peripheral blood mononuclear cells . The generation and characterization of monoclonalantibody (mAb) 8H4, specific for PrP C , and monoclonal antibody1A10, specific for DAF, have been described previously [43–45]. Pe-ripheral blood mononuclear cells (PBMC) were obtained from healthydonors by isolation over Ficoll–Hypaque density gradients followed bycentrifugation [42]. Immunofluorescent staining and PI-PLC digestions . Human PBMCwere washed with PBS supplemented with 5% newborn calf serum,0.1% NaN 3 , pH 7.4. Single cell suspensions (1  10 6 /ml) were incu-bated on ice for 45min with purified anti-PrP C mAb 8H4 or anti-DAFmAb 1A10, or an isotype control antibody. Cells were washed threetimes in medium as above and incubated for 45min on ice with 25 l lFITC-conjugated goat anti-mouse IgG (Kirkegaard and Perry Lab.,Gaithersburg, MD). Finally, samples were washed and fixed with 1%paraformaldehyde. At least 5000 cells were analyzed in a FACScan(Becton–Dickinson, San Jose, CA).For treatment with PI-PLC, cells were incubated for 30min in37  C with 60ng/ml of the enzyme in a CO 2  incubator as described [46].Treated cells were washed extensively and stained as described.Human T cell clone, NHBAc25 cells, was cultured at 2 : 5  10 5 cells/ml in complete medium (RPMI 1640, 10% pre-selected FCS,1%  LL -glutamine, and 1% antibiotics). Brefeldin A (BFA) (Sigma,MO) (36 l M/ml) was added for different lengths of time. Aftertreatment, cells were stained and analyzed by FACS as described. Allexperiments were repeated at least three times with comparableresults. Membrane fractionation and immunoblotting  . Human PBMC(2  10 7 ) were rinsed with PBS and resuspended in 1ml cold lysisbuffer [150mM NaCl, 20mM Tris–HCl (pH 6.5), 0.5% Triton X-100,2mM EDTA, 1mM PMSF, and 1 l g/ml aprotinin]. The suspendedcells were homogenized in a Dounce homogenizer, gently mixed withan equal volume of 85% sucrose in lysis buffer, and placed in thebottom of a SW40 centrifuge tube. The sample was then overlaid with2ml of 35% sucrose and 1.2ml of 5% sucrose in lysis buffer, and spun14h at 200,000  g   at 4  C. Fractions were collected in 0.4ml/aliquotsfrom the top to the bottom.For immunoblotting, proteins were separated in 12% SDS– PAGE and then transferred to Immobilon P (Millipore, Waltham,MA) for 2h at 60V. Membranes were incubated overnight at 4  Cwith mAbs and blocked for 1h at room temperature with 0.5%gelatin in PBS plus 0.05% Tween 20 and 0.05% thimerosal. Afterblocking, membranes were incubated with a 1/500 dilution of ahorseradish peroxidase-conjugated goat anti-mouse Ig antibody inPBS containing 1% BSA and 0.05% Tween 20. Bands were detectedusing enhanced chemiluminescence reagent (Amersham, ArlingtonHeights, IL). Pre-stained molecular weight markers were used asstandards. Capture ELISA for  PrP  C  and DAF  . For the PrP C capture ELISA,0.2 l g anti-PrP mAb 11G5 was coated overnight at 4  C on ELISAplates. Plates were then washed three times with PBS containing 0.1%Tween 20 (PBST), blocked for 1h at room temperature with 4% non-fat milk in PBST, and washed with PBST. One hundred  l l of culturesupernatants was added to wells and incubated overnight at 4  C.Biotinylated anti-PrP mAb 7A12 (1 l g/ml) was then added and plateswere incubated for 2h at room temperature. After washing HRP-conjugated streptavidin (BD Pharmingen, CA) was added and plateswere incubated for an additional hour at room temperature. After afinal washing with PBST, immunoreactivity was determined byadding ABTS solution (Roche Diagnostics, IN) and measuring itsabsorbance at 405nm. Recombinant human PrP was used as astandard.For quantifying DAF in the supernatant, microtiter plates werecoated with anti-DAF mAb 1A10 as described [47]. After blockingwith PBS containing 1% BSA and 0.05% Tween 20, 25 l l of superna-tants was added to wells in triplicates, and after incubation and asecond washing the amount of bound DAF was determined by  125 I-labeled anti-DAF mAb, IIH6 as described [47]. Radioactivity in eachwell was quantified in a  b  counter. Purified human erythrocyte DAFwas used as a standard. R. Li et al. / Biochemical and Biophysical Research Communications 303 (2003) 446–451  447  Results Both decay-accelerating factor and   PrP  C  are enriched inthe detergent insoluble complex in human peripheral blood mononuclear cells We used sucrose gradient flotation followed by frac-tionation and immunoblotting with anti-PrP C or anti-DAF mAb to compare the cell surface distribution of PrP C and DAF on human PBMC. As shown in Fig. 1,the localization of PrP C (upper panel) and DAF (lowerpanel) in raft (fractions 3, 4) and non-raft (fractions 9,10) sites was similar. Most of the PrP C and DAF werepresent in fraction 3, a characteristic of detergent in-soluble complexes (DIC). Most of PrP C and DAF pro-teins were therefore present in the glycolipid-enrichedmembrane sub-domains on human PBMC. Similar re-sults were obtained with the human T cell clone(NHBAc25) and a human leukemia cell line, Jurkat (notshown). Compared to DAF,  PrP  C  on human PBMC is moreresistant to cleavage to PI-PLC  The sensitivity of GPI-anchored proteins to cleavageby PI-PLC is protein and cell type-dependent [35,36].Although previous studies [2,42] have measured thesensitivity of PrP C to PI-PLC, no study has examinedthe question quantitatively by comparison to a well-characterized protein on the same cell type. We nextexamined PrP C and DAF on human PBMC with respectto their sensitivity to the enzyme. For this purpose,purified human PBMC were divided into two samples.One was treated with PI-PLC [47] and the other wasmock treated with PBS as a control. After treatment, thecells were stained with anti-PrP C mAb 8H4 or anti-DAFmAb 1A10 and analyzed by flow cytometry [42].As shown in Fig. 2, the PI-PLC treatment reduced thestaining for DAF by 75–95% ( n ¼ 5). The identicaltreatment reduced the levels of PrP C by only 25–40%( n ¼ 5).We next compared the sensitivity of DAF and PrP C on mitogen-activated PBMC to cleavage by PI-PLC. Asseen in Figs. 2C and D, PHA stimulation significantlyup-regulated the expression of both DAF and PrP C . Aspreviously found for resting PBMC, DAF (Fig. 2C) onactivated PBMC was much more sensitive to PI-PLCthan PrP C (Fig. 2D). No further reduction in the level of PrP C was achieved (1) by increasing the concentrationsof the enzyme, (2) by extending the incubation time,or (3) by repeating the PI-PLC treatment with freshenzyme (results not shown).The use of the human T cell clone, NHBAc25, al-lowed us to compare the effect of PI-PLC treatment onDAF and PrP C expression on a single cell type. Theresults were comparable to those obtained with PBMC(results not shown). The simplest explanation is that the Fig. 1. Both DAF and PrP C are present in detergent insoluble complex(DIC) in human peripheral blood mononuclear cells (PBMC). HumanPBMC were isolated from normal donors and lysed as described [42].Cell lysates were centrifuged to equilibrium on sucrose gradients. Eachfraction was then immunoblotted with either anti-PrP C mAb or anti-DAF mAb to determine the distribution of PrP C and DAF. Both PrP C (upper panel) and DAF (lower panel) are present in fraction 3 close tothe top of the gradient, suggesting that both DAF and PrP C arepresent in DIC in human PBMC.Fig. 2. DAF is more sensitive to PI-PLC than PrP C on human PBMC.Purified human PBMC were treated with 60ng/ml phosphatidylinosi-tol-specific phospholipase C for 30min in a 37  C CO 2  incubator asdescribed [42]. After treatment cells were washed extensively and di-vided into two types. One type was stained with an anti-DAF mAb andthe other tube was stained with an anti-PrP mAb 8H4. Stained sampleswere then analyzed by FACS as described [42]. At least 5000 cells weregated for analysis in each sample. Panels A and B were from restingPBMC. Panels C and D were from PBMC activated in vitro with PHAand then processed for immunostaining. Cell surface DAF is muchmore sensitive to PI-PLC than PrP C .448  R. Li et al. / Biochemical and Biophysical Research Communications 303 (2003) 446–451  GPI anchor attaching PrP C to human lymphocytes isless accessible to the enzyme but the possibility that itmay be structurally different from the GPI anchor of DAF cannot be excluded. Compared to DAF,  PrP  C  on lymphocytes has a more rapid turnover rate In the next series of experiments, we compared thehalf-life of PrP C with that of DAF. For this work weused the human T cell clone, NHBAc25, and culturedthem for different lengths of time with brefeldin A(BFA), an agent, which prevents the transport of newlysynthesized protein to the cell surface. At different timesafter culture with BFA, the cells were analyzed for PrP C and DAF expression.Histograms from one of the experiments are shown inFig. 3A. The results of another experiment in whichmean fluorescent intensity is plotted against time afterBFA treatment are shown in Fig. 3B. As seen, PrP C exhibited a much shorter half-life than DAF on thesurface of this T cell clone. Within 30min of BFAtreatment, a significant reduction in PrP C can be ob-served. In contrast, the fluorescent intensity of DAF wasunaltered throughout the entire culture period. At 3 and6h after culture the difference between PrP C and DAFexpression was even more profound. The calculatedhalf-life of PrP C was  3h and that of DAF longer than6h. The same experiment was done with the human Tcell leukemia cell line, Jurkat, and the results weresimilar (not shown). Therefore, while both PrP C andDAF are present in DIC, their turnover rates on the cellsurface differ.  PrP  C  but not DAF is shed from the cell surface The shorter half-life of PrP C on the cell surfacecould be due to a higher rate of internalization or toshedding from the cell surface. We, therefore, nextcompared the levels of PrP C and DAF present in su-pernatants of Jurkat cells cultured for different lengthsof time with BFA. We chose Jurkat cells becauseJurkat cells express comparable levels of PrP C andDAF on the cell surface (Fig. 4A). As measured usinga capture-ELISA for PrP C and DAF [45,47] increasinglevels of PrP C but not DAF were detected in the cul-ture supernatant in a time-dependent fashion. At60min after culture, only a very small amount of DAFwas detectable in the culture supernatant (Fig. 4). Theresults thus indicate that shedding of cell surface PrP C plays a role in its shorter half-life on the cell surfacethan DAF. Fig. 3. DAF has a slower turnover rate than PrP C on human T cell clone. Human T cell clone, NHBAc25 cells, was cultured in vitro with brefeldin A(BFA) (36 l M/ml) for different lengths of time. At different times after treatment, cells were isolated and stained with a control, isotype matchedMab, the anti-PrP C mAb 8H4 or the anti-DAF mAb 1A10. Stained samples were then analyzed by FACS as described [42]. Cell surface DAFappears to have a much longer half-life than PrP C . R. Li et al. / Biochemical and Biophysical Research Communications 303 (2003) 446–451  449  Discussion In this study, we found that both PrP C and DAF arepresent principally in lipid rafts on human lymphocytes.Despite their similar surface distribution, we found thatthe two GPI-anchored proteins differ in their sensitivityto PI-PLC, turnover rate, and shedding from the plasmamembrane.The reason that DAF is more sensitive to PI-PLCthan PrP C is not clear. As indicated (Results), the sim-plest explanation would be less accessibility of the PrP C GPI anchors to the enzyme. The infectious prion, PrP Sc ,is more resistancet to PI-PLC than PrP C . Masking of theGPI anchors on PrP Sc was the cause of the resistance[50,51]. It is known that GPI anchors with three acylchains are resistant to cleavage by PI-PLC [39,48] butproteins anchored by a three-footed GPI anchor aremore stable than proteins with a two-footed GPI anchor[49]. Nevertheless, PrP C exhibited a much shorter half-life on the cell surface than DAF. Characterizing thebiochemical composition of the GPI anchor on PrP C will require a much more detailed chemical analysis. Inthe cell model, PrP C molecules with pathogenic muta-tions leading to prion diseases are more resistant to PI-PLC [52]. Therefore, features other than the GPI anchorof the PrP can influence the susceptibility of the mole-cule to PI-PLC.We found that on the surface of human T lympho-cytes and Jurkat cells PrP C exhibited a half-life of    2– 3h. These findings are consistent with a recent reportstudying peripheral blood leukocytes [53]. However, ithas alternatively been reported that PrP C on the surfaceofatransfectedtumor celllinehasahalf-lifeof   6h[54].Twodifferentfactorsregulatethelevelofaproteinonthecellsurface:internalizationofthecellsurfaceproteinandrelease of the cell surface protein. Both internalization of PrP C and shedding of PrP C have been reported[53,55,56]. We found that soluble PrP C is present innormal body fluids such as, urine, milk, and blood (un-published results). Here, we reported that significantlevels of soluble PrP C are present in the supernatants of humanPBMC andJurkat cells. On theother hand,no orlittle DAF is detected in the supernatant of Jurkat cells.This finding is consistent with earlier findings thatshedding of DAF is cell type-dependent. DAF is shedfrom the surface of epithelial cells but not lymphoid cells[57]. Release of the cell surface protein can be mediatedeither by shedding or proteolytic cleavage of the protein.It has been found that released-PrP C lacks the GPI an-chor [53]. Cleavage of cell surface PrP C is unlikely to bedue to PI-PLC, as evidenced in our studies of its sensi-tivity. It is also unlikely to be due to GPI-specific phos-pholipase D (GPI-PLD) as there was no release of DAFwhich also wouldbe cleaved [58]. Shedding ofPrP C is themost likely mechanism. It also is possible that PrP C has ahigher internalization rate than DAF. It has been re-ported that PrP C is internalized by a clathrin coated pitmechanism [59].A recent study in rat neurons found that PrP C andGPI-anchored Thy-1 occupy different micro-domains onthe cell surface. These results suggest that rafts on thecell surface can be subdivided into smaller domains. It ispossible that DAF and PrP C are also present in differentsub-micro-domains on the cell surface. Differences in thesub-micro-domain could contribute to their differencesin sensitivity to PI-PLC as well as turnover rate. Suchdifferences could have a significant impact on the normalphysiologic functions of these molecules. References [1] S.B. Prusiner, M.R. Scott, S.J. DeArmond, F.E. Cohen, Cell 93(1998) 337–348.Fig. 4. PrP C has a much faster shedding rate than DAF. (A) Jurkat Tcells were stained with a control mAb, an anti-PrP C mAb or an anti-DAF mAb as described. After staining cells were then analyzed byFACS. Jurkat T cells express comparable levels of PrP C and DAF. (B)Jurkat T cells were cultured in vitro with brefeldin A (BFA) (36 l M/ml) for different lengths of time. At different times after treatment,supernatants were collected and half of the supernatant was used toquantify the amount of free PrP C by a capture ELISA as described.The other half was used to quantify the amount of free DAF asdescribed. Only PrP C is present in the culture supernatant in a time-dependent manner. A very small amount of DAF is detected at 2hafter culture. Therefore, PrP C is shed from the cell surface at a muchfaster rate.450  R. Li et al. / Biochemical and Biophysical Research Communications 303 (2003) 446–451
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks